THE INTELLIGENCE OF
HOW CHILDREN DIFFER IN ABILITY
THE USE OF MENTAL TESTS IN SCHOOL GRADING
AND THE PROPER EDUCATION OF
LEWIS M. TERMAN
PROFESSOR OF 'EDUCATION
LHLAND STANFORD JUNIOR UNIVERSITY
HOUGHTON MIFFLJN COMPANY
BOSTON NEW YORK CHICAGO
G$e Ktitaeribe f&ce&l Cambribge
TO MY GRADUATE STDDENTS
1916-1917 AND 1917-1918
THREE years ago the author of the present volume
published, in this series, The Measurement of Intelli-
gence. This represented a number of years of care-
ful scientific work, on the part of himself and his stu-
dents, in testing out and adapting to American needs
and conditions the very important foundation work
of the French scholar Binet. It was predicted at the
time that the volume would prove of fundamental
importance in pointing the way to more intelligent
school room procedure, and that in time the mental
measurement of all children not maln'rig satisfactory
educational progress would become a matter of rou-
tine in the administration of a school
The reception given to the volume, not only in this
country but in Canada and England as well, has ex-
ceeded expectations. With the entry of the United
States into the World War, and the application of
intelligence testing to our army recruits as a means of
grading capacity 'and sorting them for the serious busi-
ness of war, an entirely new impetus has been given to
intelligence measurements. As a result of the work
done in the army, as much progress has been made
during the past three years in the use of intelligence
tests as ordinarily might have been expected in a
viii EDITOR'S INTRODUCTION
In consequence, the large usefulness of intelligence
testing has beten thoroughly demonstrated, and as a
result the near future is almost certain to see the
method applied somewhat generally in schoolroom
practice to determine mental capacity and effect
proper grade classification. Intelligence testing is also
certain to play an important part in educational and
vocational guidance. The question now is not. Are
intelligence tests of value? but, How may teachers and
principals be made masters of their use? To those
responsible for the administration of both public and
private education the question has become, What
modifications of educational procedure will be neces-
sary in consequence of the new light on mental devel-
opment and school work which intelligence measure-
ments have revealed? The questions of promotions,
proper grading, types of courses of study, the skipping
of grades, over-age children, juvenile delinquency,
vocational guidance, special classes, and the proper
education of gifted and sub-normal children have all
acquired an entirely new meaning in the light of the
results which the measurement of intelligence has
In TheMeaswementcflnteMigencetihe author reduced
the tests and the procedure for giving them to simple
language, so that from a study of the book any careful
student could be trained to give them. In the present
volume he has done an even more significant thing,
viewed either from the point of view of the teacher in
EDITOR'S INTRODUCTION k
our schools or from that of the principal or superintend-
ent responsible for their administration. In this book
he has set forth in equally simple language, backed
up by the results of a number of concrete studies, and
illustrated by curves, tables, and descriptions of many
individual pupils who have been tested and studied
the educational significance of intelligence, the great
differences in the intelligence of school children, and
what may be expected from and what ought to be done
for pupils of different degrees of intellectual capacity.
The book has been written primarily for the grade
teacher, and as an introduction to the study of The
Measurement of Intelligence. Its greatest usefulness
probably will be as a book for Teacher Study Clubs
and State Reading Circles, and as an introductory
textbook for students in normal schools. A careful
study of this book by the teachers of a city or State
would contribute wonderfully to the intelligent han-
dling of children, and the study of it by prospective
teachers would open up entirely new conceptions as to
educational procedure, and would lead to a far more
satisfactory direction of the exceptional children found
in every school. The book will also prove of much
value to parents interested in the education of their
children, and especially to those whose children vary
much in either direction from the normal.
While written primarily for the teacher in service and
for parents interested in the mental development of
their children, the book ought also to prove illtiminat-
x EDITOR'S INTRODUCTION
ing to principals and superintendents of schools gen-
erally, by way of suggesting to them more intelligent
educational procedure in the classification and promo-
tion of children and the administration of courses of
ELLWOOD P. CUBBEHLEY
THIS book lias been written for the rank and file of
teachers, school supervisors, and normal-school stu-
dents. Its purpose is to illustrate the large individual
differences in original endowment which exist among
school children and to show the practical bearing of
these differences upon the everyday problems of class-
room management and school administration. It does
not treat, except incidentally, the psychological prin-
ciples underlying intelligence tests. Some of these
problems the writer has touched upon elsewhere. 1
The technique of giving the tests of the revised Binet
scale and the general significance of mental tests for
education have been set forth in some detail in another
volume of this series, The Measurement of Intelligence,*
which should be read in connection with the present
In the preparation of this volume the writer has
drawn heavily upon the data from investigations made
by a number of his students at Stanford University.
His debt to them is very great, not only for the gener-
ous way in which they have placed valuable data at
his disposal, but if possible even more for the loyalty
1 See Tennan, Lewis M., The Stamford Revision and Extension of
the Binet-Simon Measuring Scale of Intelligence. Warwick and York,
Baltimore, 1917; pp. 179.
1 Houghton Mifflin Company, 1916; pp, 862.
and enthusiasm with which they have worked to-
gether in carrying through cooperative undertakings
of the most laborious nature. What a single individual
working alone can accomplish in research with mental
tests is well-nigh infinitesimal. Substantial progress
can come only from the cooperative work of many
on closely allied problems. This volume is in large
measure the outcome of studies made by various mem-
bers of the author's classes in intelligence tests during
the years 1916-17 and 1917-18, the central topic for
each year being the relation of school success to intel-
ligence* Students who have contributed important
data to the various chapters include the following:
Virgil E. Dickson (tests of first-grade pupils);
W. M. Proctor (tests of high-school pupils);
Irene Cuneo (tests of kindergarten children);
Margaret Hopwood Hubbard (tests of superior
O. S. Hubbard (tests of fifth-grade pupils);
Isabel Preston (analysis of discrepancies between
mental age and school success);
J. K. Flanders (tests of Express Company employ-
H. E. Knollin (tests of unemployed, prisoners, and
Dr. J. Harold Williams (tests of juvenile delin-
Lowery Howard and "Virgil Dickson (tests of re-
tarded children in the schools of "X** County);
Various students who coSperated in gathering the
data on which the Stanford Revision of the Binet
. scale was based. Among these were Dr. George
Ordahl, Dr. Louise Ellison Ordahl, Grace Lyman,
Neva Galbreath, and Wilf ord Talbert.
These studies are but parts of a larger investigation
of mental growth and individual differences. Several
of them are far from complete at the time of this writ-
ing. Hundreds of children who have been tested in
the vicinity of Stanford University are being followed
up in order to discover the value of mental tests as a
means of forecasting the educational achievements
possible to children of various degrees of intelligence.
The investigation also involves the re-testing of a large
number of children in successive or alternate years in
order that typical curves of mental growth may be
established. The writer believes that studies of this
kind should entirely replace the controversial litera-
ture on the value of Binet and other mental tests.
There is no other foundation for science, whether pure
or applied, than positive, definitely verifiable facts.
Psychology is no exception.
Another study should be mentioned in this connec-
tion, although circumstances prevent the publication
of its results at present. With the assistance of a
number of Stanford University students, the group
intelligence scale devised for use in the United States
Army was given during the school year of 1917-18
to approximately six thousand pupils from the third
grade to the senior year of high school. The purpose
of the investigation was to secure data on the reli-
ability of the army tests, and to this end a large
amount of supplementary information regarding each
pupil was secured for correlation with the tests re-
sults. This information included age, grade, years in
school, nationality of parents, occupation of father,
teachers' ratings of the children on intelligence, qual-
ity of school work, and several character traits. Ap-
proximately six hundred of the same pupils had been
tested with the Stanford Revision of the Binet scale.
In every respect the results of this investigation sup-
port the data and conclusions presented in the vari-
ous chapters of this volume. The army tests, which
were given to approximately 1,700,000 soldiers, demon-
strated beyond question that the methods of mental
measurement are capable of making a contribution of
great value to military efficiency. That their univer-
sal use in the schoolroom is necessary to educational
efficiency will doubtless soon be accepted as a matter
The fact that the conclusions here offered are based
chiefly upon results secured by the use of the Stanford
Revision of the Binet-Simon tests must not be under-
stood to imply that the writer looks with disfavor upon
other intelligence scales. To the extent that the con-
clusions are valid at all, they can be confirmed by any
system of tests which affords a reasonably accurate
measure of general mental ability. However, it is not
the purpose of this volume to summarize the hundreds
of interesting and valuable investigations which have
utilized either Binet or other tests of school children.
For the most part such investigations have been di-
rected toward the improvements of methods. The
writer's present **i is the more practical one of show-
ing how the results of mental tests may be put to
everyday use hi the grade classification and in the
educational guidance of school children.
The author is indebted to Professor R. M. Yerkes
for reading several chapters of the manuscript and
for many helpful criticisms.
March 1, 1919 ,
I. SOME PRINCIPLES OF INTELLIGENCE TESTING . . 1
The Binet tests, a method of assaying intelligence
Why a mental test is significant The meaning of mental
age Mental age a basis for school grading The intel-
ligence quotient The I Q as a basis for prediction
Effect of environment on the I Q Scales for group testing.
II. AMOUNT AND SIGNIFICANCE OF INDIVIDUAL DIFFER-
Individual differences exist for all traits The causes
of individual differences in school progress Overlapping
of mental ages in the different grades The tendency to
promote by age.
IH. INDIVIDUAL DIFFERENCES AMONG KINDERGARTEN
Range in mental age Distribution of I Q's Sex dif-
ferences Significance of the tests Special need of
tests in the kindergarten The kindergarten's demands
IV. INDIVIDUAL DIFFERENCES IN THE FIRST GRADE . 4
The critical importance of the first grade A model
study of school grading Mental-age differences Men-
tal age necessary for first-grade work The influence of
age on the ability to do school work The distribution of
I Q's How the five classes differed Sex differences
Racial and social differences Correlation between intel-
ligence and other traits Predictions regarding school
progress The retarded group Feasibility of testing
all first-grade children.
V. INDIVIDUAL DIFFERENCES IN THE FIFTH GRADE . 66
Extent of the differences The two classes contrasted
Necessity of an absolute standard of comparison The
intelligence tests confirmed by other data Betardation
VI. INDIVIDUAL DIFFERENCES IN THE FIRST YEAR OF
HIGH SCHOOL 75
Age differences Mental age differences Mental age
and school marks IQ and chronological age IQ
and school work IQ and teachers' estimates of in-
telligence The relation of intelligence to elimination
Other evidence that elimination is selective Are high-
school standards too high?
VIC. THE MENTAL-AGE STANDAED FOR GRADING . . 92
Normal mental age for the different grades Sources of
error in judging school success Discrepancies between
mental age and school performance Effect of unusual
application Effect of child's personality on the teacher's
ratings Effect of timidity and lack of self-confidence
Effect of mental inertia Effects of emotional instability
or nervous tendencies Effect of home "spoiling"
Influence of physical defects Love affairs and day-
dreaming Summary The case of Margaret.
VnL MENTAL TESTS OF SCHOOL LAGGARDS . . . Ill
The number of over-age children The supposed causes
of retardation The real cause of retardation Feeble-
minded school children Grade progress of the feeble-
minded Some exceptionally difficult classes How many
children are feeble-minded? Criteria of mental deficiency
Feeble-mindedness and dullness not curable Grade
expectancy of the feeble-minded Limitations of the
special class Vocational training for backward children.
IX. THE I Q AS A BASTS FOR PREDICTION .... 135
Prediction the essence of science Limitations of pre-
diction in psychology Constancy of I Q as shown by re-
tests Personal equation of the examiner Influence of
interval between tests Influence of brightness and dull-
ness on the constancy of the I Q limits of accuracy hi
prediction of mental development Mechanical errors
as a source of discrepancy Do adenoids affect the I Q?
Curves of mental growth The I Q as a basis for pre-
dicting school progress.
X. SOME FACTS ABOUT FIFTY-NINE SUPERIOR CHIL-
Educational neglect of superior children Selection
of subjects Supplementary data I Q's Age-grade
location Teachers' ratings on quality of school work
^Educational measurements Entering age and rate
of advancement Age of learning to read Attitude
toward school work Play and recreation Trait rat-
ings Moral traits Health and physical traits So-
cial status and heredity Does the superiority tend to
XI. CASE STUDIES OF FORTY-ONE SUPERIOR CHILDREN 194
Descriptions of forty-one children of superior intel-
lectual ability Indications of superior endowment
Objections to grading superior children by mental age
Opportunity classes for superior children Class section-
ing according to mental ability.
XII. INTELLIGENCE TESTS IN VOCATIONAL AND EDUCA-
TIONAL GUIDANCE 268
Educational and vocational guidance inseparable
Limitations of vocational guidance Firemen and police-
men Express company employees Street-car em-
' ployees and salesgirls Business men Tests of college
students Tests of social and industrial failures Edu-
cational guidance The conservation of talent.
TCm. PRACTICAL SUGGESTIONS FOR THE USE OF MEN-
TAL TESTS 291
Teachers must learn to use tests Preparation needed
for Binet testing How to learn the Binet procedure
without instruction Learning to score The interpre-
tation and use of results The use of supplementary
data The Providence example The use of abbreviated
tests The vocabulary test as a brief intelligence scale
MST OP FIGURES
1. Individual differences among 145 first-grade chil-
dren in sense of humor 18
2. Individual differences among 145 first-grade chil-
dren in cheerfulness 18
3. Individual differences among 145 first-grade chil-
dren in evenness of temper . 19
4. Individual differences among 508 children in quality
of school work Id
5. Individual differences among 892 children in ability
to give sustained attention 20
6. Individual differences in height among 286 boys,
aged" 9* to 10* years 22
7. Individual differences among 1896 ten-year-old boys
in ability to win promotions in school ... 22
8. Individual differences among 88 unselected twelve-
year-old children in I Q as measured by the Stan-
ford-Binet Scale 28
9. Individual differences in intelligence among 1458
children, Grades V to VJJUL, as shown by teachers'
10. Overlapping in the mental ages of children in the
first, fifth, and ninth grades 25
1 1. Overlapping of kindergarten and first-grade children
in mental age 81
12. I Q distribution of 112 kindergarten children . . 84
18. Mental age distribution of 149 first-grade children 45
14. I Q distribution of 149 first-grade children . . 50
15. Typical "trait profile" of a very bright child . 59
16. Typical "trait profile" of a child of average intelli-
xxii LIST OF FIGURES
17. Topical "trait profile" of a feeble-minded child . 59
18. Mental-age distribution of 137 first-year high-
school pupils 78
19. I Q distribution of first-year high-school pupils . 81
20. Mental-growth curves as they would be if I Q were
21. Actual mental-growth curves of children of various
degrees of brightness 15$
22. Mental-growth curves of bright and dull children . 154
23. Mental-growth curves in two contrasting families . 155
24. Mental-growth contrasts hi the same family . . 156
25. Four exceptionally irregular growth curves , 157
THE INTELLIGENCE OF
SOME PRINCIPLES OF INTELLIGENCE TESTING*
The Binet tests, a method of assaying intelligence.
In order to find out how much gold is contained in a
given vein of quartz it is not necessary to uncover all
the ore and extract and weigh every particle of the
precious metal. It is sufficient merely to ascertain by
borings the linear extent of the lode and to take a
small amount of the ore to the laboratory of an assayer,
who will make a test and render a verdict of so many
ounces of gold per ton of ore.
A half -century ago Francis Galton predicted that it
would sometime be possible to obtain a general knowl-
edge of the intellectual capacities of a man by sinking
shafts, as it were, at a few critical points.- Already
Galton's dream is in process of realization, for in the
last decade mental testing has become one of the most
fruitful branches of psychological science. The credit
for pointing the way belongs largely to the French
psychologist, Alfred Binet, who, after more than fif-
1 For a more extended discussion of the principles of mental
testing, including detailed instructions for the use of the Stanford
Revision of the Binet-Simon Intelligence Scale, see Terman, Lewis
M., The Measurement of Intelligence. Houghton TVTifflin Company.
1916; pp. 862.
2 INTELLIGENCE OP SCHOOL CHILDREN
teen years of patient research, gave to the world in
1908 the system of mental tests now known as the
Binet-Simon Intelligence Scale. In various revised
forms the method has come into general use in public
schools, institutions for defectives, prisons, reform
schools, and juvenile courts in the United States
and in Europe. Our debt to Binet is very great, for
he succeeded in bringing psychology down from the
clouds and making it useful to men.
The Binet scale is made up of an extended series of
tests in the nature of problems, success in which de-
mands the exercise of the intellectual processes. As
left by Binet, the scale consisted of fifty-four tests,
ranging in difficulty from tests which are passed by the
average child of three years, to tests which are difficult
enough for the average adult. The Stanford Revision
has increased the number of tests to ninety and has
extended the scale far enough to measure the intelli-
gence of superior adults.
The ninety teats in the revised scale constitute an
extremely variegated series. This is necessary, since
their purpose is to measure the subject's general intel-
ligence, not his special ability in a particular line. They
include tests of memory, language comprehension,
size of vocabulary, orientation in time and space, eye-
hand coordinations, knowledge about familiar things,
judgment, ability to find likenesses and differences
between common objects, arithmetical reasoning, re-
sourcefulness and ingenuity in difficult practical situ-
PRINCIPLES OF INTELLIGENCE TESTING 3
ations, ability to detect absurdities, apperception, the
speed and richness of association of ideas, the power to
combine the dissected parts of a form board or a group
of ideas into a unitary whole, the capacity to generalize
from particulars, the ability to deduce a rule from con-
nected facts, etc. Thus the tests give a kind of com-
posite picture of the subject's general mental ability,
and since standards of comparison have been estab-
lished for each of the individual tests by trying it out
on hundreds of unselected normal children of all ages,
it is possible to express the total result of an examina-
tion in terms of "mental age" norms.
Why a mental test is significant Are we justified
in attributing real diagnostic significance to the little
intellectual "stunts" called for by an intelligence
scale? Some of these may even appear trivial. What
does it signify, for example, whether a given ten-year*
old subject names forty words or a hundred words in
three minutes? Whether he puts together the parts
of a form board in thirty seconds or in two minutes?
Whether he defines thirty words or sixty words ttf
a hundred-word list? Whether his definitions of
words are stated in terms of "use" or in terms "su-
perior to use "? Whether a series of five digits or only
a series of three digits can be repeated backwards
after a single auditory presentation? Whether there
are three, two, one, or no successes in the attempt to
draw a diamond-shaped figure from copy?
The secret lies in the standardization of the tests
4 INTELLIGENCE OF SCHOOL CHILDBEN
upon normal children of different ages. Without such
a standardization the tests would mean nothing.
Standardization is coming to play the same r61e in
psychology that it has long played in the various
branches of applied science. The architect or bridge
engineer plans his structure with constant reference
to foot-pounds of strain which various materials will
withstand. The physician analyzes a drop of blood
and, by comparison of corpuscle count and haemo-
globin with the norms for health and disease, is able
to render an important diagnosis. The psychologist
working with mental tests may be compared with the
palaeontologist who finds in a gravel bed of some
prehistoric age a skull cap, a fragment of jaw, and
a broken humerus. Although the layman might not
even recognize the human origin of such remnants, the
palaeontologist is able to tell us that the bones are
those of a middle-aged male, that the species to which
he belonged had not yet learned to stand erect, that
he probably did not know the use of fire (worn, teeth
indicate that he subsisted on uncooked foods), that his
intelligence was inferior (cranial contents only two
thirds that of modern man), and that he had prob-
ably evolved but limited power of speech (diminutive
points of attachment for the speech muscles) . A little
technical acquaintance with the standards of shape,
size, and structure of human bones has transformed
the meaningless fragments into a "missing link"
PRINCIPLES OF INTELLIGENCE TESTING 5
Perhaps no two things could be more alike to casual
inspection than the balls of two thumbs; yet one who
has been taught to read finger prints can ordinarily
find from forty to seventy separate and individually
sufficient points of identification. Just as many a
man has been hanged on the evidence of his finger
prints, so many an individual might safely be com-
mitted to an institution for the feeble-minded on the
evidence of ten or a dozen intelligence tests which
have been standardized according to age norms.
The meaning of mental age. Both the individual
tests of the Binet scale and the scale as a whole have
been standardized on the basis of age norms. The
tests themselves are located in age groups in such a
way as to bring it about that the average child of eight
years will earn by the scale a "mental age" of eight
years, the average twelve-year-old a "mental age" of
twelve years, etc. Such an arrangement was arrived
at empirically by trying out a series of tests upon hun-
dreds of normal children of different ages. The Stan-
ford Revision, for example, was based on tests of 1700
children and 400 adults.
To illustrate the use of the scale, let us suppose we
are testing a child of eight years. If our subject passes
successfully as far as the average child of eight years,
we say that his mental age is eight years, or in this
case normal. If he goes as far as the average ten-year-
old, we say that he has a mental age of ten years. If
he earns no more credit than the average six-year-old,
<j BSTTELLIGENCE OF SCHOOL CHILDREN
his mental age is six years. Binet merely took a
standard of comparison which every one uses (namely,
the standard of age) and made it definite by finding
out what intellectual performances representative chil-
dren of different ages are capable of.
It is necessary that the reader should at the outset
arrive at a correct understanding of what the term
"mental age 9 ' is and is not intended to signify. Two
misconceptions are to be avoided:
1. That each "mental age' 9 is a separate and quali-
tatively distinct level of mental attainment, contrast-
ing markedly with both the mental age which precedes
it and that which follows it. Such a use of the term is
not in harmony with the facts. Mental development
is consecutive and gradual. There is probably no
mental power, capacity, or function which has a
Minerva birth, The "faculty" in question develops
first in rudimentary form, then grows gradually
stronger and more definite until, by imperceptible
stages, it reaches a state of maturity.
2. Another misunderstanding comes from the as-
sumption that those who use the term believe a given
mental age is a stage of development which all normal
individuals pass through at the corresponding actual
age. Such a belief would imply that at the age of ten
years, for example, all children who do not belong to
some special type (defective, genius, etc.) should be
found at the ten-year mental age, eight-year children
at the eight-year mental age, etc. It is one of the
PRINCIPLES OP INTELLIGENCE TESTING 7
main purposes of this book to show how widely chil-
dren of a given age differ in mental age, and how
greatly children of adjacent ages overlap each other
in mental age.
The real meaning of the term is perfectly straight-
forward and unambiguous. By a given mental age
we mean that degree of general mental ability which is
possessed by the average child of corresponding chrono-
Mental age a basis for school grading. The signifi-
cance of mental age for the teacher lies in the fact that
it can be used as a basis for grading the pupils so as
to secure class groups of homogeneous ability. As
will be shown in succeeding chapters, the pupils of
given grades, or even the pupils of one grade in a single
classroom, are far from equal in general intelligence
or in ability to master the school work. Generally
speaking, not far from a fourth of the pupils in any
given grade have a mental level too low to make satis-
factory work in that grade possible, while another
fourth have reached a mental level which would enable
them to succeed in a higher grade.
The intelligence quotient. The mental age merely
indicates the level of development which a child has
reached at a given time. Considered apart from
chronological age it does not tell us whether a child
is bright, dull, or average. Of three children all test-
ing at the mental age of eight years, one might very
well be exceptionally superior, one average, and one
3 UNiJbiJLJULliJiiJNCJifl Utf SCHOOL CHILDREN
feeble-minded. Such would be the case if their chron-
ological ages were six, eight, and twelve years. In
addition to an index of absolute mental level, we need
an index of relative brightness. Such is the intelli-
gence quotient (I Q), which is the ratio of mental age
to chronological age. The six-year-old of eight-year
mental age has an I Q of 8/6 or 133 ; l the twelve-year-
old with a mental age of eight years, an I Q of 8/ 12, or
67. In computing the I Q of an adult subject, years
of chronological age in excess of sixteen are disre-
garded, as the development of native intelligence
seems practically to cease not far from this age.
An idea of how greatly school children differ in
brightness is shown by the analysis of the I Q's of 1000
representative children in which it was found that: 2
The lowest 1% go to 70 or below, the highest 1% reach ISO or above
44 44 44
2% " 73'
3% " " 70 *
5% - - 78
10% - 85 " ' w - 10%
15% * " 88 " " " " 15%
20% - 91 - * " * 20%
25% * 92 - * " 25%
SS% * " 95 M 83%
The intelligence quotient a basis for prediction,
ust as mental age indicates the school grade in which
, child normally belongs at a given time, so the I Q is
he basis for prediction in regard to the child's later
aental development. The possibility of such pretlic-
1 More correctly, 1.33; but the decimal point is customarily
oaitted, the quotient being understood as expressed in p<*r cent.
2 Terman, Lewis M., The Measurement of Intelligence, Uoughton
rifflin Company, 1016. See chapter v.
PRINCIPLES OF INTELLIGENCE TESTING 9
tion comes from the fact that the I Q has been found
in the large majority of cases to remain fairly constant,
at least for the ages between three or four and four-
teen or fif teen. For illustration, we will take the case
of a four-year-old child who is found to have a mental
age of five years, and whose I Q is therefore 125. The
probability is that this child will continue to have a
mental age not far from 25 per cent above his chrono-
logical age, with the consequences which may be ex-
pressed as follows:
Chronological Probable Probable school
age mental age ability
4 years 5 years Upper kindergarten .
6 " 7| " Second school grade
8 " 10 " High fourth grade
10 " 12J " Low seventh grade
12 " 15 " First year high school
It would, of course, be absurd to expect the I Q
to maintain itself at an absolutely constant figure.
Fluctuations occur for at least three reasons: (1) There
may be in exceptional cases a certain amount of irregu-
larity in the actual rate of mental development.
(8) The results of a test may be influenced to some
extent by the conditions under which it is given, the
state of the child's health, his attitude toward the
test, fatigue, and other temporary or accidental fac-
tors. Re-tests after a brief interval indicate that
errors from this source are ordinarily not large.
(3) There is inevitably a certain amount of error in
every I Q rating, due to imperfections in the scale
10 INTELLIGENCE OF SCHOOL CHILDREN
used. If the scale has been so standardized that it
yields mental ages which are too low, the I Q found
will be too low; if the scale errs in the direction of being
too generous, the resulting I Q will be too high. A
scale may err in one direction at one level and in the
opposite direction at another level. It was the most
serious fault of the original Binet scale that in the
lower range of tests it yielded mental ages which were
too high, and in the upper range mental ages which
were too low. The effect of such errors is greatly to
exaggerate the amount of fluctuation to which mental
growth is subject. It was the main purpose of the
Stanford Revision to reduce these constant errors.
Chapter XI shows in detail the degree of constancy
which may be expected for the I Q when the Stanford
Revision is used. While the law of constancy is sub-
ject to minor revisions, few things are more certain
than the essential untruth of the widespread belief
that mental development knows no regularity, and
that the dullard of to-day becomes the genius of to-
morrow. The fact is that, apart from minor fluctua-
tions due to temporary factors, and apart from oc-
casional instances of arrest or deterioration due to
acquired nervous disease, the feeble-minded remain
feeble-minded, the dull remain dull, the average re-
main average, and the superior remain superior.
There is nothing in one's equipment, with the excep-
tion of character, which rivals the I Q in importance.
Effect of environment on the I Q. The question is
PRINCIPLES OF INTELLIGENCE TESTING 11
always raised whether, in estimating a child's intelli-
gence on the basis of the I Q, it is not necessary to
make allowance for the influence of social environ-
ment. For example, it is often argued that the child
cannot know his age if he has never heard it, cannot
read and report the memory passages if he has never
attended school, cannot count from ^0 to 1 if he has
never been taught to count from 1 to 20, cannot name
the days of the week or the months of the year unless
he has heard others name them, and that therefore
the I Q can have little significance except possibly
as an index of the subject's social and educational
It is, of course, true that an individual who for his
entire life had been entirely deprived of human en-
vironment (assuming such a thing to be possible)
could not pass a satisfactory Binet test, however nor-
mal his original endowment may have been. To use
an extreme illustration, a child of ten years who had
been reared in a cage, whose wants had been supplied
while he was asleep, or by means of ingenious mechan-
ical contrivances, who had never seen a human being,
could hardly be expected to make a brilliant showing
in defining words in the vocabulary test, detecting
absurdities, repeating sentences, reading the Binet
passage, answering comprehension questions, or nam-
ing sixty words. We may go further and assume that
such a subject would be as little successful with the
three-year as with the ten-year test.
12 INTELLIGENCE OP SCHOOL CHTLDBEN
Needless to say, the Binet scale was not intended for
subjects of the type we have just described. Its use
in a given case takes for granted that the ordinary and
all but inevitable social contacts have been made, that
the subject is not deaf or blind, and that he has had
reasonable opportunity to learn the language in which
the tests are given. Children who have attended
school for any considerable time meet all of these
requirements, whatever the social status of the home.
As a matter of fact, limited acquaintance with the
language employed in the examination does not put
the subject at great disadvantage in many of the tests.
In some it does, and in testing subjects who are under
this handicap the vocabulary test and a few others
may very well be omitted. Following are two illus-
trations which show that the validity of the scale does
not hinge entirely upon the subject's knowledge of
1. Kohs tested a Belgian refugee child of nine years
who had been in America but two years. Although
this child's acquaintance with the English language
was very limited, the I Q earned on the Stanford-
Binet scale was 99. The child was also doing school
work of average quality in the fourth grade.
&. Dickson tested a Japanese boy, aged five years,
two months, who had never attended school and who
had had little opportunity to learn English; yet this
boy earned a mental age of seven years and an I Q
PBINCIPLES OF INTELLIGENCE TESTING IS
That lack of schooling does not prevent a subject
from earning an average or superior score in the test
is shown by the cases of S. S. and Gypsy Mary.
S. S. was tested at the age of seven years. He had
never been to school, and although his home advan-
tages were excellent, he had had no formal instruction
and had never learned to read. The parents believed,
perhaps rightly, that the important needs of child-
hood, apart from simple moral instruction, are food,
fresh air, and freedom for play. Nevertheless, S.
earned a mental age of ten years, eight months, and
an I Q of 153.
In 1916 a gypsy girl of sixteen years was given the
Stanf ord-Binet test in a clinic in Oakland, California.
This girl had been stolen by the gypsies when she was
about four years old, had lived with them continu-
ously until a few days before the test was made, and
had never attended a school. The I Q found was ap-
proximately 100. 1
It is not denied that the cultural status of the home
(even apart from heredity) may affect the result of the
test to some extent, although the influence has never
been accurately determined. If it were considerable,
1 TLe girl had run away from the gypsies and had told of being
kidnaped by them when a child. The gypsies denied her story and
stated that she was weak-minded and not responsible for what she
said. The mental test was given to determine her competence.
As a result of her testimony, she was freed from her gypsy parents
and returned to her home State (Montana), where she was placed
in school. Within a year she had completed the work of several
grades and was ready to enter high school.
14 INTELLIGENCE OF SCHOOL CHILDREN
we should find a marked rise of I Q in the ^ase of chil-
dren who had been removed from au inferior to a sat-
isfactory home environment. Our data on this point
are not extensive, but of a dozen or more children of
this kind whom we have re-tested, not one showed
improvement. Two such children, Walter and Frank,
have been under observation for several years. Until
the ages of five and seven years they lived in an excep-
tionally poor home. The mother was dull, the father
illiterate and a drunkard. Both of the parents died
within a year and the boys were adopted by a woman
of decidedly more than average ability who treated
them as her own sons. At the time of adoption one
tested at 73, the other at 88. Pour years later the
I Q's were 70 and 77. It is a general rule that children
of border-line intelligence improve little if at all in I Q
as they get older, notwithstanding their increased
school experience and the extra attention they receive
in special classes.
That the environment of the home affects the result
of the test but little is further shown by the fact that
occasionally in a very inferior home all of the children
except one test low, as would be expected, while that
one tests exceptionally high. In one such family
(Portuguese) there are three children who test between
76 and 88, while a brother of these tests at 130. The
latter is making a very superior record in high school,
which he entered at the age of thirteen years. The
others have not been able to complete the eighth
PRINCIPLES OP INTELLIGENCE TESTING 15
grade. All have had the same home environment and
the same educational opportunities. 1
Scales for group testing. To test each year the
intelligence of all the children by the Binet method
would involve a larger task than the school is likely
to undertake. There is accordingly a wide field for
tests which can be applied to an entire group, or class,
at onoe. The various scales have been devised for this
purpose. The group scales are given as written tests
and can be applied to an entire class of fifty or more
pupils in about an hour. To score the records requires
about ten minutes for each pupil, or a total of about
five or six hours for a class of average size. This can
be done evenings or at odd times. Most group scales
have the advantage of requiring little special psycho-
logical training either for giving the tests or scoring
them. An unfortunate limitation of such scales is that
they are not satisfactory in the lower grades where the
need for testing is greatest. As measures of intelli-
gence they are probably somewhat less accurate than
scales for individual testing, but their obvious advan-
tages make them deserving of wide use with pupils
of the upper grades and high school. 2
However, no group scale will ever do away with the
necessity of individual testing. Rather it mokes the
need for individual testing more obvious. All the
* The mental growth curves .of two of these children are shown in
Fig. 24, p. 156.
1 The Otis Group Tests and the instructions for using them are
supplied by The World Book Company, Yonkers, New York.
16 INTELLIGENCE OF SCHOOL CHILDREN
pupils in the fourth grade and beyond should be given
a test by the group method every year, and those
whose scores are either very high or very low in the
group examination should be given a Binet test. As
will be shown later (chapter IV), it is highly desirable
that every pupil be given a mental test within the first
half-year of his school life.
AMOUNT AND SIGNIFICANCE OF INDIVIDUAL
Individual differences exist for all traits. When
many unselected children of a given age are examined
for any trait, large individual differences are found.
This is true whether the trait in question is height,
weight, strength, lung capacity, number of blood cor-
puscles, hearing, vision, intelligence, courage, con-
scientiousness, social adaptability, vanity, or any one
of a hundred others. Figures 1 to 5 illustrate typical
individual differences among school children in sense
of humor, cheerfulness, evenness of temper, quality of
school work, and ability to give sustained attention.
The graphs show the per cent of pupils who were classi-
fied by their teachers as "very inferior," "inferior,"
"average," "superior," or "very superior" in regard
to each of the traits.
The above graphs represent the distribution of
teachers' ratings; that is, estimates based on personal
observation of the pupils rated. Actual measurement
of the traits would have been preferable to ratings,
had such measurement been possible, but there arc
1 For a more extended discussion of individual differences see
E. L. Thorndike, Educational Psychology, vol. ni (1914), pp. 141-
308. Teachers College, Columbia University, New York City.
18 INTELLIGENCE OF SCHOOL CHILDREN
FIG. 1. INDIVIDUAL DIFFERENCES AMONG 145 FIRST-GRADE CHIL-
DREN IN SENSE OF HUMOR. (Teachers' ratings.)
FIG. 2. INDIVIDUAL DDTFERENCES AMONG 145 FIRST-GRADE Cini/-
DBEN IN CHEERFULNESS. (Teacher* 1 ratings.)
Very inferior Inferior Average Superior Very superior
5.6% 17.2% 49.6% 20% 7.6%
FIG. 3. INDIVIDUAL DIFFERENCES AMONG 145 FIRST-GRADE CHIL-
DREN IN EVENNESS OF TEMPER. (Teachers' ratings.')
Very inferior Inferior Average Superior Very superior
6.2% 17.9% 61% 22.1% 3.8%
FIG. 4. INDIVIDUAL DIFFERENCES AMONG 503 CHILDREN IN
QUALITY OP SCHOOL WORK. (Teadiers 9 ratings.)
INTELLIGENCE OF SCHOOL CHILDREN
Very inferior Inferior Average Superior Very nuperior
4.6% 21.2% 30.8% 27% 7.4%
FIG. 5. INDIVIDUAL DIFFERENCES AMONG 392 CIULDRENT, FIRST TO
EIGHTH GRADE, IN ABILITY TO GIVE SUSTAINED ATTENTION.
still many domains of mind and character for which
measuring scales have not been devised.
It may be argued that the individual differences
represented in the above graphs arc spurious; that
they merely reflect the varying degrees of error in the
judgments of those who furnished the ratings. It is
an extremely significant f ax&, however, that whenever
we succeed in devising a method for actually measuring
a mental trait, as large individual differences are found
for it as for such physical traits as height or weight.
The latter are of course susceptible of as accurate
measurement as practical purposes are likely to de-
mand. The progress of children through the grades
of a school system can be measured in terms of age-
grade status with sufficient accuracy. In the case
INDIVIDUAL DIFFERENCES 21
of intelligence, also, the individual differences can be
measured; perhaps less accurately than height, yet far
more accurately than they can be estimated on the
basis of common observation. Figure 6 illustrates
typical differences among ten-year-old boys in height;
Figure 7, typical differences among ten-year-old boys
in ability to win promotions in school; Figure 8, typi-
cal differences among twelve-year-old boys in I Q; and
Figure 9, typical differences among 1458 children as
shown by teachers' ratings for intelligence. Attention
is directed to the fact that individual differences are
equally in evidence for the four traits.
The causes of individual differences in school prog-
ress. In the case of a physical trait such as height,
perhaps few would deny that the differences found
represent in the main differences in original endow-
ment. That the progress children make through the
grades of a school system is also chiefly dependent
upon original endowment is neither so obvious nor so
generally believed. The common opinion seems to be
that nearly all children are capable of satisfactorily
accomplishing eight grades of school work in eight
years, and that if they fail to do so it is because of
faulty school management. The remedies most often
proposed for the prevention of retardation are better
attendance laws, school census reform, extension and
improvement of medical inspection, flexible grading,
and adaptations of the course of study.
That reform in all these lines is needed, for other
2 INTELLIGENCE OP SCHOOL CHILDREN
119-123 124-128 129-133 134-138 139-143 144-148 149-153
2.1% 11% 22% 29.2% 23.7% 10.2% 1.7%
FIG. 6. INDIVIDUAL DIFFERENCES IN HEIGHT AMONG 236 BOYS, AGED
9J TO 10} YEARS. (Baldwin.)
Grade I II m IV V VI VH
FIG. 7. INDIVIDUAL DIFFERENCES AMONG 1896 TEN-YEAR-OLD
BOYS IN ABILITY TO WIN PROMOTIONS IN SCHOOL. (Shows grade
location cf ten-year-old boys in Salt Lake City, May, 1916.)
I Q 06-75 76-85 80-05 96-105 100-115 116-125 126-135
5% 15% 20.5% 28% 19.5% 11% .8%
FIG. 8. INDIVIDUAL DIFFERENCES AMONG 83 UNSELECTED 12-YsAR-
OLD CHILDREN IN I Q AS MEASURED BY THE STANFORD-BINET SCALE
Very inferior Inferior Average Superior Very superior
1.5% 16.2% 63.3% 15.5% 3.6%
FIG. 9. INDIVIDUAL DIFFERENCES IN INTELLIGENCE AMONG 1458
CHILDREN, GRADES V TO VIII. (Teachers 9 ratings.)
34 INTEUIGENCE OF SCHOOL CHILDREN
reasons as well as for the reduction of retardation, will
be admitted by all. We are beginning to learn, how-
ever, that all of these measures combined are power-
less to reduce greatly the number of over-age children
in the grades. Notwithstanding the persistent cam-
paign which has been waged against the evils of retar-
dation for the last dozen years, the number of retar-
dates remains to-day much the same as it was when
the campaign began. We are justified in raising the
question whether the most important cause of retarda-
tion has been located, and whether it is one that can
In the various chapters of this book certain data
from intelligence tests will be analyzed in the attempt
to formulate an answer to the above question. The
facts which will be presented point fairly definitely to
the conclusion that the differences which have been
found to exist among children in physical traits are
paralleled by equal differences in mental traits, par-
ticularly intelligence. It will be shown that these in-
nate differences in intelligence are chiefly responsible
for the problem of the school laggard; that the so-
called "retarded" children on whom we have expended
so much sympathy are in reality nearly always above
the grade where they belong by mental development;
and that the real retardates are the under-age children,
who are generally found from' one to three grades
below the location which their mental development
would warrant. In other words, it will be shown that
INDIVIDUAL DIFFERENCES 25
the retardation problem is exactly the reverse of what
it is popularly supposed to be.
Overlapping of mental ages in the different grades.
The extent of the school's failure to grade children
according to their ability will be evident from an
examination of Figure 10, which shows the actual dis-
tribution of mental ages disclosed by the Stanford-
Binet scale in the first grade, the fifth grade, and the
first year of high school in typical public school sys-
tems of California. It will be seen that not only do
the first-grade children greatly overlap those of the
fifth grade, and fifth-grade children those of the first
year of high school, but that the brightest child in the
first grade has all but reached a point in mental ability
corresponding to that of the lowest pupil in the high
school. The brightest of the fifth-grade pupils is
above the median mental level for the first year of high
4 6 6 7 8 9 10 11 12 13 U 15 16 17 IS 19
FlG. 10. OVERLAPPING IN THE MENTAL AGES OP CHILDREN IN THB
FIBST. FIFTH, AND NINTH GRADES
36 INTELLIGENCE OF SCHOOL CHTLDBEN
school, and the brightest of the first grade reaches the
median for the fifth grade.
That there are children in the first grade as old
chronologically as the youngest in the eighth grade is
generally understood and deplored, but few teachers
are aware of the fact that mental ages are scattered
through the grades hardly less promiscuously. Table
1 shows the grade distribution by mental age of 676
III IV V VI VII VIII
7-6 to 8-5
8-6 to 9-5
9-6 to 10-5
10-6 to 11-5
11-6 to 12-5
12-6 to 13-5
13-6 to 14-5
TABLE 1. DISTRIBUTION IN TEE GRADES OP 676 UNBELBCTED
By "7-6" is meant 7 years and 6 months. This form of expression will be used
throughout in both text and tables.
unselected pupils below the high school who are men-
tally eight years old or older.
INDIVTOUAL DIFFERENCES 27
The failure of school grading to give groups of homo-
geneous chronological age is a matter of hardly any
importance compared with its failure to give groups
of homogeneous mental ability. The chronologically
old and the chronologically young may and often do
belong together, the mentally old and the mentally
young do not. Notwithstanding the sifting which
takes place at the end of each school year, the resulting
classification of children has been so far from success-
ful that, generally speaking, the lowest 25 per cent of
pupils in any grade belong mentally in a lower grade
and the highest 25 per cent in a higher grade. Only
the middle half are classified approximately where
they should be. Usually more than 15 per cent are
at least two grades removed from the one in which
they belong by mental age.
The tendency to promote by age. It was stated in
an earlier part of this chapter that the grade progress
of the school child is governed largely by original
endowment. However, facts such as those just pre-
sented show that endowment is by no means the sole
factor; for if it were, children would be more correctly
graded according to ability. The other factor is the
persistent tendency of teachers to promote by the cal-
endar. The dull are allowed to become somewhat re-
tarded, but are nevertheless promoted beyond their
ability to do the work. Occasionally the brightest are
allowed to become accelerated, but comparatively
rarely, and almost never as much as they deserve.
28 INTELLIGENCE OF SCHOOL CHILDREN
Here arc two children, both in the fifth grade, who
offer a typical illustration:
A. Boy; age 14-8; mental age 8-6; I Q 60; years in school
seven and a half; quality of school work "very inferior";
grade status on the usual basis of reckoning, retardation of
three years. In reality this boy is accelerated two years, for
his mental level of 8 J is at least two years below that neces-
sary for satisfactory work in the high fifth grade.
B. Girl; age 9-8; mental age 13-1; I Q 130; years in
school three; quality of school work "very superior"; grade
status reckoned on the usual basis, two years acceleration.
This girl is not really accelerated, but retarded, for her
mental level of 13 years would enable her to do average work
hi the seventh grade.
The one criterion of fitness for promotion should be
ability to meet the requirements of the next higher
grade. Pupils of the type of Child A, kept always at
tasks that are hopelessly beyond their ability, never
learn the meaning of success. Those like Child B
miss the mental and moral stimulus which comes from
intense application to tasks commensurate with abil-
ity. We see how badly misplaced any measure of
reform would be which was designed merely to "pre-
vent retardation/' in the usual sense of that term.
But what is the solution of the problem of over-age
children? Are they to be required to repeat more
grades than they now do? Would not the policy of
rigidly holding these children in the grade correspond-
ing to mental age be even more discouraging than the
present practice of over-promoting them? It would
INDIVIDUAL DIFFERENCES 29
be unfortunate indeed if we were obliged to choose
between the two evils. Perhaps another solution is
possible if we will only cease to think exclusively in
terms of cross-section education. Instead of a single
curriculum for all, merely divided into eight successive
levels, it would be better to arrange parallel courses of
study for children of different grades of ability. Some
such solution seems necessary if we are to adjust school
work to the abilities of the children and at the same
time avoid the admittedly serious evils of repetition.
INDIVIDUAL DIFFERENCES AMONG KINDERGARTEN
TESTS were made by Miss Cuneo of 112 children
attending five kindergarten classes in the cities of San
Jose and San Mateo, California. The majority of the5
pupils came from middle-class homes, a few from each
extreme of the social scale. All were American born.
The ages were as follows:
Range in mental age. Although the total range of
actual ages was from 3$ to 7 years, all but 19 of the
pupils were between 4 and 6. As will be seen from
Table 8, the range in mental age was greater than this;
namely, from SHfc to 7-7. Of the 112 pupils, 5 were
mentally below 4 years, 85 between 6 and 7, and 3
above 7. The kindergarten group all but overlaps in
1 Written with the assistance of Irene Cuneo.
AMONG KINDERGARTEN CHILDREN 81
mental age the fifth-grade group described in chap-
ter V. The highest mental age found in the kindergarten
was 7-7, the lowest in the fifth grade 7-9. The chances
are that if twice as many had been tested an actual
overlapping would have been found.
TABLE 2 . MENTAL AGES OF 112 KINDEEGAETBN CHILDREN
7* 8 8* 9 9*
FIG. 11. SHOWING OVERLAPPING OP KINDERGARTEN AND FIRST-
GRADE CHILDREN IN MENTAL AGE
Comparison of the mental ages of these 113 kinder-
garten children with the mental ages of 150 unselected
first-grade children tested by Dickson may be made
from Figure 11. Nearly a fourth of the kindergarten
82 INTELLIGENCE OF SCHOOL CHILDBEN
children equaled or exceeded the median mental age
of those in the first grade, and more than half equaled
or exceeded the lowest fourth of first-grade children.
A large proportion of these kindergarten children have
a mental maturity which would enable them to do
satisfactory work in the first grade.
"The most abrupt break in the curriculum is that
from the kindergarten to the first grade. At all other
points every effort is made to bridge the gaps. The
transition from first grade to second, from fifth to
sixth, etc., is almost imperceptible. Even the first
year of high school is rapidly being integrated with the
last year of the grammar school so as to give the child
an unbroken educational path which he may traverse
from the first grade to the university. The kinder-
garten alone holds aloof, worships at the shrine of a
special methodological cult, and treats its children as
belonging to a different order of human beings.
The tests of Dickson and Cuneo show how little
justification there is for such an attitude. The fact
that nearly a fourth of kindergarten children do not
differ at all in mental ability from average first-grade
children, and that a fourth of first-grade children are
on a par with the median kindergarten child, indicates
that it would be well for the teachers of these two
grades to come to some kind of understanding.
Distribution of intelligence quotients. The I Q's
of the 112 children are listed below in order from high-
est to lowest.
AMONG KDSTDERGARTEN CHTLDBEN 33
152 114 106 93
146 114 105 93
142 114 103 92
136 113 103 91
130 113 103 91
130 113 102 90
129 113 102 90
126 112 102 90
126 112 102 90
125 111 102 88
124 111 101 86
124 110 101 86
123 110 100 85
122 110 100 85
121 110 100 85
121 109 99 84
121 109 98 82
121 109 98 81
121 109 98 80
120 109 97 80
119 108 97 80
119 108 96 80
118 107 96 79
117 107 96 77
117 107 96 76
114 107 94 75
114 106 94 72
114 106 93 61
The range is from 61 to 152, that is, from feeble-
mindedness to very unusual superiority. While only
one could be certainly classed as a defective, there are
at least three others who are at the borderline of men-
tal deficiency. The lowest 25 per cent fall to 91 or
below, the highest 5 per cent reach 117 or above.
The median is 106.
84 INTELLIGENCE OF SCHOOL CHILDREN
Figure 12 shows graphically the number falling in
the I Q groups 60-69, 70-79, etc.
60-70-80-90-100- 110- 120- 130- 140- 150-
w 79 89 99 109 119 129 139 149
.9% 4.6% 11.6% 19.6% 26% 20.5% 12.5% 2.6% 1.8% .9%
FIG. 12. I Q DISTRIBUTION OF 112 KINDERGARTEN CHILDREN
Sex differences. Is there a sex difference in intelli-
gence at the kindergarten age? We have asked many
kindergarten teachers this question and have often
received an affirmative answer. The opinion seems to
prevail that girls, even at this early age, are somewhat
more precocious than boys. Comparison of Miss
Cuneo's 65 boys and 47 girls suggests that this opinion
may not be without foundation. The medians and
upper and lower quartiles were as follows:
AMONG KINDERGARTEN CHILDREN 35
Although the brightest subject tested was a boy,
the median for girls is five points higher than for boys.
This is not a large difference, but it is appreciable. It
is probable that, age for age, girls are slightly superior
to boys in the kind of intellectual ability measured by
the usual type of intelligence test. This conclusion is
borne out by the results of many other investigations
by the test method. It is also in harmony with sex
comparisons based on teachers' ratings and school
marks. In the present study, 56 of the boys and 47
of the girls were rated for intelligence and 46 of the
boys and 36 of the girls for quality of school work.
The results were as follows :
TEACHERS' RATINGS ON INTELLIGENCE
TEACHERS' RATINGS ON SCHOOL WOBK
36 INTELLIGENCE OF SCHOOL CHILDREN
Although the superiority of the girls in the tests is
very slight, sometimes almost negligible, we have
found in something like a dozen separate studies that
for a given age or grade the girls invariably make a
significantly better showing than boys when rated by
iheir teachers either for intelligence or for quality of
school work. We do not attempt to say whether girls
make a better use of their intelligence or whether they
are more responsive and so appear brighter than they
are. Both causes may enter.
Significance of the tests. What do the large indi-
vidual differences revealed by the tests signify in
terms of future educational achievement? Is it pos-
sible, as a result of a forty-minute test of a child who is
only four or five years old, to forecast with any degree
of assurance his educational career? With accuracy,
no; in general terms, yes. There is little likelihood
that the child who tested at 61 1 Q will ever go above
median nine-year or ten-year intelligence. On the
other hand, the three brightest children in the group,
who tested as high as 140, could in all probability be
made ready for high school by the age of eleven or
Miss Cuneo re-tested seventy-seven of her pupils,
(1) Twenty-five pupils, interval of two days; ()
twenty-one pupils, interval of half-year; (3) thirty-one
pupils, interval of two years. The agreement between
the first test and the repeated test was very dose
AMONG KINDERGAETEN CHILDREN 37
except in a few cases. The median amount of change
in I Q was only six points. One fourth of the subjects
showed a change of three points or less, and one fourth
eight points or more.
Of the thirty-one pupils re-tested after two years,
there were four who had earned an I Q of 130 or above
in the first test. Two years later all still tested above
130 and all were doing "superior" or "very superior"
work in the second grade. Three of them had gained
an extra promotion. Of the ten at the other extreme
who had earned an I Q of 100 or less, not one had
gained an extra promotion.
However, there is a serious source of error to guard
against when testing children of this age. Kinder-
garten children are in the bashful stage and are likely
to respond only with silence to tests which they could
easily pass. The examiner must, therefore, take care
to get into rapport with the child if he would avoid the
error of mistaking diffidence for lack of intelligence.
Special need of tests in the kindergarten. There is
one reason why tests are more necessary in the kinder-
garten than anywhere else, if the intellectual differ-
ences which exist among pupils are to be discovered.
In other school grades the work itself constitutes a
kind of intelligence test. The first-grade child who
cannot learn to read, or the fourth-grade child who
cannot learn long division, is readily recognized as
inferior. The work of the average kindergarten offers
no such clear-cut criterion of intellectual normality.
38 INTELLIGENCE OF SCHOOL CHILDREN
The games, drawing, sand-pile activities and card-
board construction may disclose certain differences,
but these are vague and lack meaning.
This difficulty was reflected in teachers* ratings of
* he kindergarten children for intelligence. When
i jsked to estimate the intelligence of each child on the
usual scale of five: very superior, superior, average,
inferior, and very inferior, the teachers protested that
there was almost nothing in kindergarten work on
which they could base a judgment. The ratings on
intelligence were finally secured, but they correlated
with I Q's only to the extent of .29. This is not more
than half the correlation usually found in the grades
above the kindergarten. The correlation of I Q's and
the ratings for quality of school work was only .27,
and that between mental age and ratings for quality
of school work only .43.
Certain disagreements between I Q's and ratings
were due to failure to take account of age differences.
The 6-year-old is rated "superior" on an I Q about
10 to 20 points lower than a 4^-year-old must have to
reach this class. The former gets into the "superior"
class as easily with 100 1 Q as does the latter with 115
or 120. The following cases illustrate this error:
G. J., age 3-8, mental age 4-2, I Q 113, was rated as
"inferior." M. L. 9 also testing at 113, but aged 5-9, was
N. W., age 3-11, mental age 4-3, 1 Q 109, is rated "in-
ferior." J. M., age 6-11, mental age 6-6, 1 Q 94, is rated
AMONG KINDERGARTEN CHILDREN 39
M . S., age 4-1, mental age 5-4, 1 Q 130, is rated "inferior."
J. P., age 6-3, mental age 6-10, I Q 109, is rated "very
The kindergarten's demands upon intelligence.
The correlation between mental age and quality of
work is shown in Table 3.
Mental age -
7- 4 to 7-10
6-10 to 7- 3
6- 4 to 6- 9
5-10 to 0- 3
5- 4 to 5- 9
4-10 to 5- 3
4- 4to4- 9
8-10 to 4- 3
3- 4 to 3- 9
TABLE 3. CORRELATION BETWEEN MENTAL AGE AND QUALITY OF
SCHOOL WORK FOR 80 KINDERGARTEN CHILDREN
It will be seen that although there is a marked ten-
dency for those of low mental age to do "inferior**
work and for those of high mental age to do "superior "
work, children of every mental age from 8 J to 7| are
rated as doing "average" work. This can only mean
that the activities of the kindergarten do not make
very serious demands on general mental ability.
Is this a legitimate ground of criticism of the pre-
primary curriculum? The answer will depend upon
one's philosophy of the kindergarten. There is much
to be said for the kindergarten as a combination of
40 INTELLIGENCE OP SCHOOL CHILDREN
nursery and a place for socialized play. Prom this
point of view its purpose is to serve the life of instinct
and emotion, rather than to nourish intelligence. It
is a point of view which is attractive, because it has
the support of sentiment. Certainly no one would
wish to see children of four or five years harnessed to
intellectual work to the exclusion of play. This, how-
ever, is not the only alternative. Perhaps some of the
kindergarten activities could be so adapted as to neg-
lect less than they now do the appeal to intelligence.
Possibly a half -hour of the day could be advanta-
geously reserved for work of a somewhat more intel-
lectual character than the hopping, skipping, sand
digging or "busy work" of the typical conservative
kindergarten, much of which has little to commend
it from any point of view. Madame Montessori's
injection of more serious activities into the kinder-
garten would seem to mark a distinct advance.
The contrast between the usual work of the kinder-
garten and that of the first grade could be justified
only on the theory that the child of four or five years
is all instinct and emotion, and that suddenly at the
age of six or seven he is brought by a sudden meta-
morphosis into the life of intellect. A comparison of
the intelligence tests of kindergarten and first-grade
children shows how untenable such a theory is.
But if adjustments are due on the part of the kinder-
garten, they are perhaps just as much needed in the
first grade. The chasm must be bridged from both
AMONG KINDERGARTEN CHILDREN 41
sides. The mere fact that a child has passed his sixth
birthday is not a sufficient justification for robbing
him of all the freedom he has enjoyed in kindergarten
INDIVIDUAL DIFFERENCES IN THE FIRST GRADE
The critical importance of the first grade. The first
grade is the most critical in the school system. It is
the place above all others where the raw material with
which the school is to work should be correctly evalu-
ated. Success or failure for the child's school career
hangs often upon his success or failure in the first grade.
In a way school administrators appreciate this fact.
Effort is usually made to place the best teachers in
charge of the entering pupils. School doctors, school
nurses, and school dentists are commonly urged to
give special attention to the younger children. Never-
theless, it is in the first grade that retardation scores
its worst record. In the average city approximately a
fourth of ihe pupils Jail of promotion at the end of the
Schools for backward children ordinarily do not
draw from classes below the third grade. By this time
the dull pupil is already a lost cause. Special classes
for superior children, when they exist at all, are too
likely to confine their efforts to bright pupils whose
intellectual progress has already been retarded by
several years spent in the educational lockstep. By
1 Written with the assistance of Virgil E. Dickson.
DIFFERENCES IN FIRST-GRADE CHILDREN 4S
this time their intellectual ardor has cooled and the
edges of their mental faculties have been dulled.
We cannot deny that the task which faces first-
grade teachers and supervisors is very difficult. Every
twelve months the schools of the United States receive
something like three million fresh recruits. What we
are urging is the immediate assaying of all this ma-
terial, year after year. The task is difficult, but we
believe it is worth while.
A model study of school grading. The facts which
will be presented are from an investigation by Dickson,
who has analyzed the results of nearly a thousand
Stanford-Binet tests of first-grade children. This ac-
count will include the results of the first 150 tests only.
The group included all the pupils found in the first
grade in five different schoolrooms in the vicinity of
Stanford University. The rooms will be designated
by the letters A, B, C, D, and E. Room A included
chiefly pupils of Spanish and Italian descent, but
American born. Room B represented a mixture of
races, with American predominating. Room C was
similar to A, but contained a considerable number of
Portuguese, in addition to a few Italian and Spanish.
Room D drew mainly from the upper and middle
social classes of American extraction. The children
of Room E were all from an exceptionally well-to-do
Dickson used the tests only as a point of departure
for a more intensive study of the pupils. His purpose
44 INTELLIGENCE OP SCHOOL CHILDREN
was the practical one of trying to translate individual
differences into terms of classroom management. Ac-
cordingly, the cooperation of the teachers was enlisted
in the collection of a large amount of supplementary
data regarding each child, including:
1. Date of entering school;
8. Occupation of the father;
3. Nationality of each parent;
4. The teacher's estimate of the child's intelligence;
5. The teacher's rating of the quality of the school
6. The teacher's rating of each child (on a scale of
five, as in the case of intelligence and school work) on
each of the following twenty-four traits: power to give
sustained attention, persistence, social adaptability,
leadership, initiative, evenness of temper, emotional
self-control, physical self-control, will power, cheerful-
ness, courage, sense of humor, obedience, conscien-
tiousness, dependability, intellectual modesty, unself-
ishness, coSperativeness, speed, industry, personal
appearance, popularity among fellows, self-expression,
Mental-age differences. The ages extended from
5 years 7 months to 11 years, a range of nearly 5|
years. The mental-age range is even greater; namely,
from 3 years to practically 11 years. The highest
mental age among these first-grade pupils considerably
overlaps the lowest we have found in the eighth grade.
The number at each mental age is shown in Figure 13.
DIFFERENCES IN FIRST-GRADE CHILDREN 45
2% 15.4% 20.8% 84.2% 21.5% 4% L8% .67%
FIG. 13. MENTAL-AGE DISTRIBUTION OF 149 FIBST-GBADB
Mental age necessary for first-grade work. Chil-
dren are expected to start to school at the age of six
years. The assumption is that the degree of mental
maturity corresponding to this age is necessary for
successful work in the first grade. Is this assumption
justified? The question can be answered by a com-
parison of the quality of school work done by the chil-
dren of different mental ages. The agreement be-
tween mental age and the ratings for quality of work
done is shown in Table 4.
It is evident from this table that mental age is a
fairly good index of a child's ability to do the work
of the first grade. No child below the mental age of
6 years is rated as doing school work above the aver-
age, while of the 22 pupils rated as "very inferior'* in
46 INTELLIGENCE OF SCHOOL CHILDREN
Quality of school work
9-0 to 9- 5
8-6 to 8-11
8-0 to 8- 5
7-6 to 7-11
7-0 to 7- 5
6-6 to 6-11
6-0 to 6-5
5-6 to 5-11
5-0 to 5- 5
4-6 to 4-11
to 4- 5
TABLE 4. SHOWING HOW QUALITY OP WORK IN THE FIRST GRADE
DEPENDS UPON MENTAL AGE. (Correlation .725)
school work, all are below the mental age of 6. The
agreement, however, is not perfect. Of the 41 children
who are mentally 7 years or above, 7 .are rated as doing
"inferior" work. All but 2 of these are enrolled in
rooms D and E, where the average mental age of the
pupils is unusually high. The inference is that teach-
ers have judged these pupils by too high a standard of
performance. If placed in room A probably any one
of them would rank average or above.
On the other hand, 13 of the 98 pupils below the
mental age of 6 years are rated as doing school work
of " average " quality. A satisfactory explanation was
found in every case. Two of the 18 were repeaters,
DIFFERENCES IN FIRST-GRADE CHILDREN 47
one of whom was 10 years old and had been in school
over two years. Under the circumstances they would
naturally be expected to do average work, even though
they are slightly below the mental level of 6 years
The other 11 were all enrolled in classes A and B, in
both of which the average mental age was extraordi-
narily low. In room A, 13 of the 38 pupils were below
the mental age of 5 years, and in Room B 18 out of 39.
The 11 pupils whose ratings are in question averaged
in mental age 5 years and 10 months. It is therefore
not surprising that their school work should have
been rated as average in an inferior class.
From such data as the above, collected from all his
1000 cases, Dickson concludes that below the mental
age of 6 years the child is not fully ready for the first
grade, and that below the mental age of 5 J years the
chances that really standard first-grade work will be
done are practically negligible. We are beginning to
see why a fourth of the pupils in the first grade fail of
promotion, for Dickson finds 38 per cent below the
mental age of 6 years, and 7 per cent below 5 J.
On the other hand, there are in the first grade many
pupils of the other extreme of ability who are kept at
work which is too easy to command their best efforts.
Of the 150 children, 15 (10 per cent) are above the
mental age of 7 years. All of these, and in addition
perhaps half of the 26 who tested between 7 and 7J,
could quickly be made ready for the second grade.
By the present regime these are injured no less than
48 INTELLIGENCE OF SCHOOL CHILDREN "
The influence of age on the ability to do school work.
The one condition that the school imposes upon those
who would enter is that they shall have passed a given
birthday. If the age criterion were adequate we should
invariably find the best records in the first grade made
by the oldest pupils, and the poorest records by the
youngest pupils. The reverse is the case. Of the
89 pupils who were above the chronological age of 7J
years, only 3 were rated above average in quality of
school work; of the 10 who had reached the age of 8|
years, none. Conversely, of the 83 who were rated
above average in quality of work, 18 were under 6|
years and 80 were under 7| years, chronologically.
This finding is not new. Every one who has given
mental tests to any considerable number of school
children has found that the best pupils in a given grade
are almost invariably the youngest; the poorest pupils,
the oldest. In the present instance this is true, not-
withstanding the fact that most of the older pupils are
taking the work for the second, third, or fourth time.
Some, in fact, had started to school before their young-
est classmates were born.
Does age give any advantage whatever apart from
the degrfee of mental maturity which has been attained,
or does school success depend entirely upon mental age,
except in so far as it is influenced by such extraneous
factors as industry, illness, emotional instability, etc.?
Dickson attempted to answer this question by
comparing the quality of school work done by older
DIFFERENCES IN FIRST-GRADE CHILDREN 49
(dull) pupils and the younger (bright) pupils of the
same mental age. For the comparison he took two
groups of pupils, all of whose mental ages were be-
tween 6 and 7 years. The pupils of one of these groups
were also chronologically between 6 and 7 years, with
I Q's ranging between 96 and 105. The pupils of the
other group were between 8 and 9 years chronologi-
cally, with I Q's ranging from 7 to 84. The average
mental age in the two groups was almost exactly the
same. The comparison gave for the older group an
average rating of 3.7 in school work; for the younger
group, an average rating of 3.12. Since on our five-
step scale of rating, 3 means "average" and 2 means
"superior," it is seen that there is a difference of more
than a half -step in favor of the younger group. The
significance of this finding is enhanced by the fact
that the older group had attended school for an av-
erage period of 1.9 years, the younger for an average
period of less than one year. The older group has
two years the advantage in age, with all the inciden-
tal experience which age brings, and in addition the
advantage of a year more of school attendance; nev-
ertheless their work is less satisfactory than that of
the younger pupils who are at the same level of men-
tal age. We have found this to be the rule in a
number of similar comparisons. The additional spon-
taneity and adaptability of young normal pupils
slightly outweigh the advantage of the additional
experience and schooling which older pupils of the
same mental age have enjoyed.
50 INTELUGENCE OF SCHOOL CHILDREN
The distribution of I Q's. The range of I Q's for
the 150 pupils is 45 to 145. The-median I Q is 88, and
it is interesting to note that the brightest child and the
dullest child are about equidistant from the median.
The lowest 5 per cent fall to 77 or below, the highest
25 per cent reach 184 or above. Figure 14 shows the
distribution of intelligence quotients grouped in ranges
of ten: 40 to 49, 50 to 59, 60 to 69, etc.
40-60-60-70-80-90- 100- 110- 120- 130- 140-
49 69 69 79 89 99 109 119 129 139 149
.7% 4.7% 8,7% 18.8% 20.1% 14.1% 16.8% 8.7% 4.7% 1.3% 1.3%
FIG. 14. I Q DISTRIBUTION OF 149 FIBST-GBADE CHILDREN
The low median I Q might be due either to real aver-
age inferiority of the pupils tested or to a defect of the
scale causing it to yield mental ages too low. That the
latter is not the true explanation is indicated by the
high average I Q (107) earned by Miss Cuneo's kinder-
garten children, who were tested on almost exactly the
same part of the scale. 1 All the supplementary data
1 Stanford-Binet tests of more than 2300 children in the kin-
dergarten and primary grades of Council Bluffs, Iowa, yielded an
average I Q of 99 points, which is within one point of that which
would be expected if the scale were correctly standardized.
DIFFERENCES IN FIRST-GRADE CHILDREN 51
confirm the test results in showing that in three of the
five rooms, enrolling 107 of the 150 pupils, there was
an excessive number of children of low mentality.
Had only rooms D and E been examined, the scale
would have seemed to err in the direction of too great
Of the 150 children, 1 tested below 70 I Q and 12
below 60. Those below 60 may safely be considered
feeble-minded* and probably a majority of those be-
tween 60 and 70. Most of the low cases were in two
It will be recalled that the teachers were asked to
estimate the intelligence of each pupil on the scale of
1, 2, 3, 4, 5. The extent to which these estimates
agreed with the tests is shown in Table 5.
jtimates of in
TABLE 5. AGREEMENT BETWEEN THE TEBTS AND TEACHBES*
ESTIMATES OF INTELLIGENCE. (Correlation .79)
52 INTELLIGENCE OF SCHOOL CHILDREN
The correlation in Table 5 (.79) is fairly high. By
a painstaking analysis of individual cases of disagree-
ment, Dickson was able to show that most of these
were caused by the failure of the teachers to take the
child's age into account. This is an error which we
have found over and over, one from which it seems
impossible for teachers to free themselves even when
expressly cautioned to do so, as they were in this case.
The child of 8 years who has a mental age of 6 years
(I Q 75) and is doing almost average work in the first
grade, is likely to be rated not far from average in in-
telligence. The teacher forgets that an average child
of 8 years ought to be doing average work in the high-
second or low-third grade.
How the five classes differed. These five teachers
are expected to accomplish the same work, to turn out
a similar product at the end of the year. A comparison
of the material with which they are working shows
that any such expectation is impossible of realization
and unfair to the teachers and to the children. Tables
6 and 7 show how greatly rooms A, B, and C diffei
from rooms D and E in the mental ages represented.
Median I Q
; below 5*
TABLE 6. SIGNIFICANT DIFFERENCES IN THE MENTAL COMPOSITION
OF FIVE FmsisGBADE CLASSES
DIFFERENCES IN FIRST-GRADE CHILDREN 53
, 60- 64
TABLE 7. I Q DISTRIBUTION IN THE FIVE CLASSES
The average mental age in room E is fully two years
above that in room B, and the median I Q 36 points
higher. The average child of room E excels the aver-
age child of room B in brightness as much as an aver*
age normal child of 100 1 Q excels a feeble-minded child
of 66 1 Q. Room A has three pupils who grade feeble-
minded, room B anywhere from seven to thirteen,
room C two to four, room D possibly one, and room E
none. Rooms A and C have no pupils who test as
high as 120, and room B only one; but 1.5 per cent
of the pupils of room D and 28.6 per cent of those in
room E grade this high or higher. These differences in
54 INTELLIGENCE OF SCHOOL CHILDREN
relative endowment are reflected in the number of
repeaters found in the five rooms.
A third of the pupils in room A and half of those in
room B are incapable of doing standard first-grade
work. They are not doing it. The lack of progress
on the part of the pupils in room B was so evident that
the teacher was in despair and the superintendent
doubted her efficiency. But there was nothing wrong
with the teacher. Her task was simply impossible.
On the other hand, half the pupils of room E have
reached a level of mental development which would
enable them to do the work of the second grad,e; three
or four of them, the work of the third grade. The lot
of this teacher is a happy one. Her pupils are able to
learn without instruction.
When a class is so far above or below the average in
ability we would expect the teacher to be aware of the
fact. That these teachers were only partially so is
shown by the distribution of their ratings on school
work in the different classes. Although the teacher in
room B correctly rates the work of her pupils very low,
the teacher of room D rates (incorrectly) more of her
pupils below than above average. Plainly her stand-
ard is too high, and it is so because the average menta f
age in her room is above seven years. The teacher oi
room E complained that six of her fourteen pupils
were not doing what she would consider good work.
The average mental age in this class was above 7$
years. Dickson estimated that "average" work in
DIFFEBENCES IN FIRST-GRADE CHILDREN 55
class E was in reality better than "superior" work in
any of the other classes.
Sex differences. Of Dickson's 150 subjects, 79
were boys and 71 were girls. Only a small difference
was found in the I Q's of the sexes, the girls having the
advantage of three points in median. However, as
rated by the teachers for quality of school work, the
girls made decidedly the better showing. This was
true even when the comparison was between boys and
girls of the same mental age. Among the 150 pupils
were 15 boys and 11 girls who tested between 95 and
105 I Q. The ages, mental ages, and the I Q's of these
two groups were almost exactly the same, yet the boys
secured an average rating of 3.4 (.4 below "average");
the girls an average rating of 2.81 (.19 above "aver*
age")- That is, boys of average intelligence may be
expected to do less than average school work, girls of
average intelligence to do better than average school
work. As stated by Dickson, "It may be that the
school curriculum is better adapted to the needs and
interests of girls; that girls excel in industry and appli-
cation; that girls more willingly submit to direction in
a task; that girls are better behaved than boys and
that school marks are influenced by deportment; that
teachers (all women) are better suited to teaching
girls than boys; or it may be any one or a combination
of many of the causes that might be mentioned."
Racial and social differences. Three of the five
rooms (A, B, and C) contained a large element of
56 INTELLIGENCE OP SCHOOL CHILDREN
Spanish, Portuguese, and Italian children, also a
number of North-European parentage. Most of these
were born in the United States and all spoke English.
The median I Q's were as follows:
Race Number Median I Q
Spanish 37 78
Portuguese 23 84
Italian 25 84
North-European ... 14 105
American 49 106
The children were classified as to social status ac-
cording to the occupation of the father. The classifica-
tion was based upon Taussig's division of occupations
into five "non-competing" groups: (1) professional;
(2) semi-professional or higher business; (3) skilled;
(4) semi-skilled; and (5) unskilled. The correlation
between I Q and the ratings on social status was
found to be .48. The median I Q or classes 4 and
5 taken together was 82.5; for classes 1 and 2 taken
together, 112.5. Only one child in class 5 tested above
115, and only one in classes 1 and 2 below 85. Two
thirds of these in classes 1 and 2 were above 100, and
seven eighths of those in classes 4 and 5 below 100.
However, bright children do occur in the lower occu-
pational groups, and when they do they stand out by
Correlation between intelligence and other traits.
It will be recalled that as a part of the supplementary
data Dickson had the teachers rate their children on
DIFFERENCES IN FIRST-GRADE CHILDREN 57
twenty-four mental and moral traits, in addition to
intelligence and school work. The interest in such
ratings lies in their bearing on the debated question
whether good traits tend to go together or whether
superiority in certain lines is likely to be offset by
inferiority in others. The latter belief is called "the
theory ^>f compensation." It is commonly thought
that the possession of a number of undesirable traits
is almost certain to be compensated by marked superi-
ority in other traits. Every one knows that this is
sometimes true in individual cases. If it were the rule,
however, there would be negative correlations among
some of the traits, and a given individual would show
a great deal of unevenness in the ratings received.
Such negative correlations were not found. The
traits listed below all correlated positively with intelli-
gence and with one another. The correlations with
intelligence are shown in Table 8 in order of amount,
beginning with the highest.
It is interesting to note that sense of humor, power
to give sustained attention, persistence, and initiative
all correlate highly with intelligence. They probably
depend in large measure upon intelligence. The cor-
relation of social adaptability with intelligence is also
high, indicating that there is little truth in the theory
that bright children tend to be socially queer or out-
casts. The low correlation of obedience, unselfish-
ness, and emotional self-control with intelligence are
58 INTELLIGENCE OF SCHOOL CHILDREN
Trait t with 1 Q
1. Sense of humor 58
2. Power to give sustained attention 54
3. Persistence 53
4. Initiative 53
5. Accuracy 52
6. Will power 50
7. Conscientiousness 48
8. Social adaptability 47
9. Leadership 44
10. Personal appearance. 44
11. Cheerfulness 43
12. Cooperation 43
IS. Physical self-control 42
14. Industry 40
15. Courage 39
16. Dependability 38
17. Self-expression (speech) 37
18. Intellectual modesty 34
19. Obedience 34
20. Popularity among fellows 34
21. Evenness of temper 31
22. Emotional self-control 29
23. Unselfishness 29
24. Speed 28
TABLE 8. CORRELATION BETWEEN I Q AND TEACHERS' RATINGS
ON VARIOUS MENTAL AND MORAL TRAITS
As a rule, the ratings given an individual child ran
fairly uniform through the list of traits. 1 Pew children
showed numerous oscillations from high to low rat-
ings. Bright children as a rule were rated "superior "
1 The precaution was taken of having each teacher rate all her
children on one trait before taking up the next trait.
Ro. 15. TYPICAL " TBAIT PROFILB " OF A VERY BBIGHT CHILD.
(I Q 133, average rating, 1.34)
FIG. 16. TYPICAL "TRAIT PROFILE " OF A CHILD OF AVERAGE
INTELLIGENCE. (I Q 103, average rating, 3.04)
FIG. 17. TYPICAL " TRAIT PROFILE " OF A FEEBLE-MINDED
CHILD. (I Q 45, average rating, 4.46)
; ; : .' .' : : i : i .' : : I <
IlIII^^Ii^I'WM"'" j> ^*-^
* ' ' ' &
_ <3 Z
rH Ii-eOOf-<^<M I "
rH ; : ;fHcoco ;?H ;
GO-*I-H<M : : w
coH<iH( : 2
Q* t-l * CO
" J 2
,H -4 rH H Ol CO CO CO CO * * ^P
DIFFERENCES IN FIRST-GRADE CHILDREN 61
or "very superior" in nearly all traits, occasionally
dropping to av< -age. Very dull children as a rule were
rated "inferior" or "very inferior" in nearly all traits,
occasionally rising to average. Figures 15, 16, and 17
reproduce from Dickson's report sample "profiles" of
one bright, one dull, and one average pupil.
So marked was the correlation between I Q and
average rating on the twenty-four traits that knowing
only the average of these ratings one could roughly pre-
dict what the I Q would be. As shown in Table 9, the
correlation is .76. Of the ten pupils with an average
rating above 2 (1 being highest and 5 lowest), not one
tested below 110 1 Q. Of the sixteen with an average
as low as 4, none tested as high as 100, and only one as
high as 90. Conversely, of the six pupils testing 125
or above, none had an average rating below 2.24; and
of the ten pupils testing below 65, none had an aver-
age rating as high as 3.24.
Predictions regarding school progress. After the
tests had been made and all the supplementary data
had been analyzed, Dickson made a prediction re-
garding each child's probable school progress. Fol-
lowing are samples of these predictions:
Child No. 1. Age 5-10; mental age 6-10; I Q 117; school
work 2; low-first grade; in school one-half year. Forecast:
Work should continue superior. Should finish fourth grade
in three to three and a half years.
Child No. 15. Age 6-4; mental age 5-7; I Q 88; school
work 3; low-first grade; in school 1 year. Forecast: Child
68 INTELLIGENCE OF SCHOOL CHILDBEN
is dull and quality of school work will go lower. Present
rating of 3 is explained by low average standard in the
class. Will probably lose one year before completing fourth
Child No. 21. Age 8-2; mental age 6-8; I Q 81 ; school
work 4; low-first grade; in school 2 years. Forecast: Very
dull, and probably will not finish fifth grade by age of four-
teen years. Not suited to the regular course of study.
Child No. 86. Age 10-0; mental age 5-1; I Q 51; school
work 5; high-first grade; in school S| years. Forecast: This
child will never reach the fourth grade. Should be placed
in an institution for the feeble-minded.
Child No. 122. Age 7-6; mental age 10-11 ; I Q 145; school
work 1; high-first grade; in school $ year. Forecast: Should
be coached on essentials and moved ahead. May be ex-
pected to complete the fourth grade within 2f years after
entering school and is capable of doing so in 1 J years.
Child. No. 117. Age 7-3, mental age 7-4; I Q 101 ; school
work 4; high-first grade; in school 1 year. Forecast: This boy
is average normal, and . an average class would do satis-
factory work without repeating. However, as the average
I Q of his class is 112, he is likely to lose a half year or more
before completing the fourth grade.
It is easy enough to make predictions. How well
time will justify them is another question. They will
be checked annually as long as the children can be
followed. Thus far they have been checked up once,
a year after the tests were made. The findings were
in the large majority of cases in perfect agreement
with the forecast. In a number of cases in which the
forecast was not borne out, it could be shown that
either: (1) the original supplementary information fur-
DIFFERENCES IN FIRST-GRADE CHILDREN 63
nished by the teacher was incorrect (age misstated,
etc.) ; or (&) the teacher had erred in promoting a child
who should not have been promoted, or in holding a
child back who should have been allowed to go on.
The retarded group. Of the 150 pupils, 83 were
retarded, according to the Ayres standard. Dickson
examined the data regarding these 33 children with a
view to discovering the causes of their retardation.
The retarded group was divided into three classes:
(1) those showing late entrance; (2) those who entered
at normal age but progressed slowly; and (3) children
showing both late entrance and slow progress. The
facts regarding these three classes are stated by Dick-
son as follows:
1. Late entrance. "Of the five children who show late
entrance, only one has normal mental ability (I Q 99). She
has made regular progre- since entering school and is now
doing work of average quality. The remaining four are sub-
normal children mentally, belonging either in the feeble-
minded or in the border-zone group. The very fact that
these four children have a low mental level is the most prob-
able cause of their retardation."
2. Normal entrance, slow progress. "Eighteen children
show entrance at normal age, but slow progress. All of
these are repeaters, several for the third or fourth time. The
mental level of each one is low. Four are probably feeble-
minded, the rest would classify in the border-zone or in the
dull-normal groups. Only one (I Q 91.7) even approaches
the average normal mental level."
3. Both late entrance and slow progress. "Ten children
show both late entrance and slow progress. Eight of these
have low mentality; one has a mental level approaching the
64 INTELLIGENCE OF SCHOOL CHILDREN
normal (91 1 Q) and is now doing average work; one has a
normal level (97 I Q) and the facts at hand do not suggest
any cause of retardation."
"Qf the 83 retarded children, only 2 have normal mentality
as shown by the testa. Stated in another toay, 98.9 per cent of
all the retardation in these five rooms is found in children of low
mental level. While there may be contributory causes, low
mentality is undoubtedly the chief cause of retardation in these
five rooms of first-grade children. 99
Feasibility of testing all first-grade children. The
first task of the school when it gathers its newcomers
together should be to give each child a mental test to
determine the nature of his endowment. The ' :st
should then be checked up by a large amount of sup-
plementary data and by an annual appraisement of
Granting the desirability of giving every child in
the first grade ,a mental test, is it possible to do so?
One is tempted to answer that it is possible because it
is necessary. Each teacher may very well test her
own pupils, or, if it is preferred, all the testing may be
done by a few teachers who have had special training
for such work. The latter plan has been followed
in Providence, Rhode Island, where during a single
school year almost 1000 first-grade pupils were tested
by teachers specially detailed for the task. It is re-
ported that the experiment has been a great success;
that the tests showed clearly why a third of their pu-
pils were f ailing of promotion in the first grade. As
a result of the experiment the course of study for the
DIFFERENCES IN FIRST-GRADE CHILDREN 65
first year was differentiated, and the mentally imma-
ture pupils were given work of a pre-primary nature.
In the schools of Council Bluffs, Iowa, all the pupils
of the kindergarten and primary grades are tested,
and entrance to the first grade from the kindergarten
is based entirely upon mental age.
INDIVIDUAL DIFFERENCES IN THE FIFTH GRADE *
HUBBABD investigated the amount of heterogeneity
in the classes from the fourth to the eighth grades of
the Alameda public schools. The study took into
account age, race, sex, social status, achievement as
shown by school marks, educational measurements
and teachers' ratings on intelligence. Data were
secured from approximately 2000 pupils. Such wide
ranges of individual differences were found that it was
decided to select two classes in the fifth grade for more
intensive study by means of the Stanford-Binet tests.
The two classes were chosen at random. One enrolled
38, the other 41 pupils.
Extent of differences. The mental ages, ages, and
I Q's found in the two classes are shown in Table 10.
The individual differences revealed in Table 10
are enormous and startling. Class A has an age
range from 9| to almost 14 years; class B, from 9|
to almost 15 years. This would not be so serious if
all were of approximately equal mental ability. Such
equality was conspicuously lacking. The pupils of
class A ranged in mental age from less than 10 yean
to more than 15 years; those of class B, from 7f years
to 14 years. Two pupils of class A have reached a
1 Written with the assistance of O. 8. Hubbard.
DIFFERENCES IN FIFTH-GRADE CHILDREN 6?
TABLE 10. AGES, MENTAL AGES, AMD I Q's OF PUPILS IN Two
68 INTELLIGENCE OF SCHOOL CHILDREN
stage of mental development corresponding to that of
the average pupil in the first year of the high school.
Either of these pupils could by six months' appropriate
training be prepared to do good high-school work. In
the normal course of events they will not reach high
school for three and a half years. At least two pupils
in each class are mentally equal to eighth-grade work.
Almost exactly half of the pupils of class A are men-
tally ripe for promotion to the sixth grade, at least,
and one fourth are ready for the seventh grade. Two
months of appropriate training would doubtless fit
them for such promotion.
On the other hand, class A contains eight pupils and
class B fourteen pupils who are at the fourth-grade
level of mental development (9| to 10| years); while
Class B contains seven who probably belong in the
third grade. One of the latter, a pupil of 13 years, has
a mental age below 8 and belongs more nearly in the
high second grade. Taking the two classes together
we find all levels of ability represented from that nor-
mal in the second grade to that normal in the first year
of high school.
Consider for a moment the contrast between the
brightest and dullest pupil in each class. In class A
the highest I Q is 148; the lowest, 78. In class B the
highest is 144; the lowest, 60. A child in the 140 1 Q
class should be able to attain marked success in one
of the learned professions, but all the refinements of
educational method are incapable of bringing a child
DIFFERENCES IN FIFTH-GRADE CHILDREN 69
of 60 I Q to the level of seventh-grade ability. If
both should remain in school the former will be win-
ning Phi Beta Kappa honors at college graduation
while the latter is still struggling with simple fractions
or long division. The difference between 140 1 Q and
60 1 Q is 80 points. The difference between an average
child and a high-grade idiot, who will never develop
beyond three years, is also about 80 points. In the
former case, we do not think of the contrast as being
so great because our perception of intellectual differ-
ences in the upper ranges is much less acute than for
the lower ranges. We are on guard against stupidity;
,we often fail to recognize superiority.
Table 10 shows the usual relationship found between
chronological age and mental age in a given grade.
Low mental age goes with high chronological age, and
low chronological age with high mental age. Express-
ing it differently, the lowest I Q's are possessed by the
oldest pupils, the highest I Q's by the youngest. In
class A the third highest mental age belongs to the
next to the youngest pupil. In class B the second
highest mental age is that of the youngest pupil.
The two classes contrasted. Considered as two
groups which are expected to cover the same work in
a given time, class A and class B present a striking
contrast. In class A, 44 per cent of the I Q's are 110
or above; in class B, only 10 per cent. In class A, 19
per cent of the I Q's are below 90; in class B, 44 per
cent. In class A the median I Q is 108, in class B, 91.
70 INTELLIGENCE OF SCHOOL CHILDREN
The median mental age in class B is slightly over 10
years; that in class A, slightly under 12 years. In
other words, the median mental ability of class A cor-
responds to that normal to the sixth grade; the median
of class B to that normal for the fourth grade. Ten
per cent of the pupils of class B test below 70; in class
A, none. In class B only 10 per cent of the pupils test
as high as 110; in class A, 49 per cent. As would be
expected, the two classes presented an entirely differ-
ent picture. The pupils of class A were interested,
alert, and above the average in industry; those of
class B inert and unresponsive.
From what we know of the significance of the I Q
the educational possibilities of these pupils can be
predicted with a fair degree of assurance. Approxi-
mately 15 or 20 per cent of the pupils of class B will
never, with any amount of instruction, be able to do the
work of the eighth grade satisfactorily, and 50 per cent
are too inferior in endowment ever to complete a
four-year course in the average American high school.
Of class A, close to 80 per cent should be able to gradu-
ate from a high school.
Necessity of an absolute standard of comparison.
The teachers of these two classes were not far from
equal in ability, training, and devotion to their work.
Both were above average. Surely, one would suppose,
they must have been keenly aware of the intellectual
composition of their classes. They were not, except
in the vaguest sort of way. Each teacher knew that
DIFFERENCES IN FIFTH-GRADE CHILDREN 71
she had some bright and some dull pupils. How
bright or how dull was not known. Each teacher
could rate her pupils only by comparing them with
others in the same class. The teachers' classification
of the pupils into the usual five groups, very superior,
superior, average, inferior, and very inferior, gave the
Class A Class B
(per cent) (per cent)
Very superior 2.6 0.0
Superior 15.7 5.2
Average 63.1 76.3
Inferior 15.7 13.1
Very inferior 2.6 5.2
Although the number rated above average is much
larger in class A than in class B, as it should be, the
number rated below average is exactly the same in the
two classes, namely, 18.3 per cent. The teacher of
class B did not know that her pupils averaged nearly a
year below fifth-grade ability, nor did the teacher of
class A know that her pupils averaged a year above.
Neither teacher suspected that her class covered a
range of four or five years in mental ability. In both
classes the significance of over-ageness and under-
ageness had been overlooked, for over-age pupils had
been consistently rated too high and the under-age
pupils too low by each teacher. It had not occurred
to the teacher of class B that a high-school education
was out of the question for half of her pupils. Even
if she had stopped to consider the fact that several of
72 INTELLIGENCE OF SCHOOL CHILDREN
her pupils were three or four years over-age for the
grade, how could she have known that this retardation
would not be made up later?
The intelligence tests confirmed by other data. The
contrast between the two classes as shown by the intel-
ligence tests is confirmed by the Courtis tests, the
Stone Reasoning test, and the Ayres Spelling lest.
Following are the median scores of each class in these
tests and the median scores in the same tests for all
the fifth-grade classes of the city taken together.
Class A Class B Entire city
Addition 7.20 3.16 4.98
Subtraction 8.12 2.80 4.81
Multiplication 5.43 1.75 3.56
Division 3.66 0.00 2.63
Reasoning 2.56 1.72 1.73
Spelling 80. 73.75 74.10
In the four fundamentals and in reasoning the aver-
age difference between the classes amounts to more
than two grades. The difference in spelling is consid-
Retardation and acceleration. If we use the Ayres
standard and call a pupil retarded who is in the fifth
grade and 12 or more years old, then a fourth of the
pupils of class A and 29 per cent of the pupils of class B
are retarded. This is a rather liberal standard. If
we use 11J years instead of 12 as a basis for figuring
retardation in the fifth grade, the amount is increased
to 41.5 per cent in class A and to 87 per cent in class B.
DIFFERENCES IN FIFTH-GRADE CHILDBEN 78
The normal mental age in the fifth grade is 11 years.
Below 10J a pupil cannot ordinarily be expected to do
satisfactory work. Of the eleven over-age pupils in
Class B (i.e., over 12 years of age), only two are as
high as 12 years, mentally, and seven are below 10$
mentally. Five are at the third-grade level of mental
ability. On the basis of mental age not. a .single pupil.
in this class is retarded, but seven are accelerated.
Five of these are accelerated fully two years. The
median I Q of the eleven over-age pupils is. 74.
Of the ten pupils in class A who are over-age (above
12), only three are as much as 12 mentally. Of the
remaining seven, five are correctly located according
to mental age and two are a full grade accelerated.
Again we see that the chief cause of retardation is not
irregular attendance, the use of a foreign language in
the home, bad teeth, adenoids, malnutrition, etc., but
inferior mental endowment. Educational reform may
as well abandon, once for all, the effort to bring all
children up to grade.
We have just seen that the over-age pupils were, on
the basis of mental age, really accelerated. Turning
now to the under-age pupils, we find that these are
the real retardates. Class B has one pupil and class A
has two pupils who are less than 10 years in chrono-
logical age. Their mental ages are 14-3, 18-10, and
12-8. Two of the three are mentally ripe for the eighth
grade, the other for the seventh. The "accelerates "
are in fact badly retarded. It is always so. We can-
74 INTELLIGENCE OF SCHOOL CHILDREN
not too often repeat that the retardation problem is
exactly the reverse of what it is commonly supposed
to be. On the basis of chronological age, class A, with
24.4 per cent of its pupils above 12 years, makes a
better showing than class B, with 29 per cent above
12 years. On the basis of mental age, however, 49 per
cent of the pupils of class A are retarded, as contrasted
with 10.5 per cent of class B.
From the point of view of mental hygiene the con-
ditions in the two classes, while different, are almost
equally unsatisfactory. Class A has 49 per cent above
the standard mental age for the grade; class B has 76.5
per cent below. The 49 per cent of class A find the
work too easy; most of the 76 per cent of class B have
a constant struggle to keep then- heads above water.
For both conditions, the educational lock-step, with
its tendency to promote by the calendar, is responsible.
Reform will have to be based upon a consideration of
individual differences measured by mental and educa-
INDIVIDUAL DIFFERENCES IN THE FIRST YEAR OF
HIGH SCHOOL i
NOTING that a third or morejof the pupils who enter
high school do not remain to begin the second year of
work, Proctor decided to attack the problem at its
most critical point by investigating the abilities of
first-year students. The aim of the study was (1) to
find how greatly first-year pupils differ in intelligence;
() to trace the dependence of school success upon
intelligence as measured by the tests; (3) to find what
relation exists between intelligence and elimination;
and (4) to investigate the possible value of intelligence
tests in educational and vocational guidance.
All the pupils who entered the Palo Alto, California,
High School during the school year 1916-17 were given
a Stanford-Binet test. The number in this group was
107. The testing was continued the following year,
in part with the Stanford-Binet and in part by the
use of a modified form of the Otis Group Scale. Alto-
gether, intelligence measurements were made of ap-
proximately 850 first-year pupils in seven different
high schools, and of 250 pupils in the eighth grade.
The purpose of the study was not merely to discover
individual differences, but also to discover what bear-
1 Written with the assistance of W. M. Proctor.
76 INTELLIGENCE OF SCHOOL CHELDKEN
ing these have upon educational guidance. It was
therefore necessary to check up the test results in as
-many ways as possible. Supplementary data secured
at the time the tests were given included nationality,
age, school marks in all the subjects, vocational ambi-
tion, occupation of father, and teachers* estimates of
intelligence. What was still more important, Proctor
followed up the cases over a period of two years, in
order to note any changes that might occur in the qual-
ity of the school work and to correlate school success
with the test results. 1
As typical of his findings we will present in this
chapter some of the results of the Stanf ord-Binet tests
of 137 pupils who had just entered the Palo Alto High
School. In every respect the results secured by the
Binet tests were closely paralleled by the group meas-
urements of more than 700 additional pupils.
Age differences. The age range was from 13-0 to
19-3, -with a median of 14-11. The median age for a
thousand unselected pupils entering New York High
Schools was 14-5; the median for 1042 in Iowa City,
14-9. While these age differences are of interest, they
do not necessarily furnish ground for criticism of school
grading. Par from maintaining that children ought
to be graded more by age than they are, it is one of
the main purposes of this book to show that grading is
to analyze the material with which high schools work. Professor
Proctor's book should be in the hands of every high-school teacher.
DIFFERENCES IN HIGH-SCHOOL PUPILS 77
based too much upon age. As will be seen presently,
the oldest of thesWpupils are mentally far below the
ability necessary for success in the first year of high
school, while the youngest are invariably retarded one
or more grades below the level of their mentality.
Mental age differences. In our discussion of mental
ages of high-school pupils it is necessary to point out
that mental ages secured by the Stanford-Binet above
14 or 15 years have something of an arbitrary mean-
ing. No one knows exactly what median intelligence
is for the ages 15, 16, 17, etc., because it is prac-
tically impossible to secure unselected subjects above
14 years. By that age the pupils of inferior ability
begin to drop out of school. Accordingly, when we
speak of the mental age, 16, 17, etc., we aie using
these figure$.rather as scores than as mental ages in
the literaf sense. We simply know that 16 denotes a
higher mental level than 15, 17 higher than 16, 18
higher than 17. With this understanding, however,
we will continue to employ the term mental age as in
The mental age scores of the 137 high-school fresh-
men ranged from 12-8 to 19-6, the latter being the
highest possible on the Stanford-Binet. The lowest
was earned by a girl whose chronological age was 19-3,
the highest by a boy whose chronological age was 13-8.
Figure 18 shows the per cent of total number at each
Only 4 of the 137 pupils were below the mental age
78 INTELLIGENCE OF SCHOOL CHILDREN
of 13 \ years. All of these were over-age ("retarded")
pupils. It appears that but for the tendency of teach-
ers to promote on the basis of age, rather than on the
FIG. 18. MENTAL-AGE DISTRIBUTION OF 137 FIRST-YEAR HIGH-
basis of ability, there would be few if any pupils in
this high school much below the mental level of 14
years. On the other hand, of the 31 pupils who have a
mental age score of 17 or above, 26 are less than 15|
years of age. The median mental age of the 137 pupils
The highest mental age in Hubbard's fifth-grade
classes (15-3) not only overlaps those of the first-year
high school, but almost reaches the median for the
latter. However, the lowest mental age in the high-
school group (12-8) is not nearly as low as the median
for the fifth grade.
If we consider the mental age 14-5 to 15-5 to be
that normal for the first year of high school, then 27,
DIFFEBENCES IN HIGH-SCHOOL PUPILS 79
or 20 per cent, are mentally below the standard mental
age for the grade, and 80, or 58 per cent, are above.
There are 41 pupils, or SO per cent, above the mental
age of 18J; and 88, or 16 per cent, above the mental
age of 17$. It could perhaps hardly be maintained
that all of these 22 ought to be doing the work of the
junior year, as this work is now constituted, but one
is tempted to raise the question whether high-school
curricula are not framed for a higher level of mental
ability than is justifiable.
Mental age and school marks. Is success in high
school largely determined by mental age, as was found
to be the case in the first and fifth grades? The answer
will be found in Table 11, which shows the correlation
between mental age and average school mark for the
111 pupils who are still in school.
18} to 14}
14} to 15}
15* to 16}
16} to 17}
17} to 18}
18} to 10}
TABLE 11. RELATION BETWEEN SCHOOL MASKS AND MENTAL AGE
80 INTELLIGENCE OP SCHOOL CHILDREN
The correlation is moderately high, but considerably
lower than is found in the grades below the high school.
The following facts are, however, very significant:
1. Of the 5 pupils with an average mark of A, not
one is below the mental age score of 17 years..
2. Of the 28 whose average score is B + , not one is
below the mental age of 15 years.
3. Of the 56 who earned a mental age score as high
as 16| years, only 8 have an average mark below B.
4. Of the 12 with a mental age below 14 , 8 earned
an average mark of C or lower.
5. The only pupils tested whose mental ages were
below 13J years (four in number), had already been
eliminated because of failure, and so do not appear
in Table 10.
Throughout Proctor's study it appears that the
standards of work which are maintained in the first
year of average California high schools cannot be
satisfactorily met by pupils with a Stanford-Binet
mental age below 13 years, and that below the mental
age of 14 years the chances of success are not good. In
rare instances the pupil of 12-year mental age is able
to m J passing grades, but only by virtue of excep-
tional application and an attractive personality.
Intelligence quotients. For the group of 107 pupils
entering in September, 1916, the I Q's ranged from 79
to 136, with a median of 105. The lowest 25 per cent
fell to 96 or below, the highest 25 per cent reached 117
or above. The median for the boys was 107; for the
DIFFERENCES IN HIGH-SCHOOL PUPILS 81
The distribution of I Q's is shown in Figure 19. The
most striking thing about the distribution is that only
three cases appear below 85, and only eight cases be-
low 90. Above 90 the number of cases increases with
marked suddenness, indicating that entrance to this
high school is pretty well barred to children who test
much below 90.
70-79 80-89 90-99 100-109 110-119 120-129 180-139
.9% 6.6% 29% 23.3% 21.4% 14% 4.6%
FIG. 19. I Q DISTRIBUTION OF FIEST-YBAB HIGH-SCHOOL PUPILS
Except for the smaller number in the lower range,
the distribution of I Q's of first-year high-school pupils
is similar in form to that found for the lower grades.
However, the Stanf ord-Binet probably grades a trifle
severely at "the upper end. As is shown elsewhere
an I Q of 130 in the case of a child of 15 years is
probably equivalent to an I Q of 140 for a child under
12. Even so, the range of I Q's from 79 to 138 is very
82 INTELLIGENCE OF SCHOOL CHELDKEN
I Q and chronological age. There was found, as
would naturally be expected, a high negative correla-
tion (-.74) between I Q and chronological age, which,
of course, simply means that the children who enter
high school young are generally brighter than those
who enter late.
. B t
TABLE 12. SHOWING NEGATIVE CORRELATION BETWEEN AGE
AND I Q. (Correlation -.72)
As shown in Table 12, no pupil below 13} years
tested lower than 120. Of the 30 pupils below 14}
years of age, not one tested lower than 100, and only
lower 'than 110. It is evident that to enter this high
school on schedule time ordinarily requires decidedly
better than average intelligence. On the other hand,
of the 38 pupils who were above the age of 15}, only 11
tested as high as 100, and only 2 as high as 110. These
38 pupils constitute the retarded group, again indicat-
ing that the chief cause of retardation is mental inferi-
ority. Of the 38, 70 per cent are below 100 I Q. As
DIFFERENCES IN HIGH-SCHOOL PUPILS 83
we have already stated, the lowest I Q was that of a
girl who was over 19.
The negative correlation between age and bright-
ness is further illustrated by the scores made in the
vocabulary test. Table 13 shows that in general the
largest vocabularies are possessed by the youngest
pupils, the smallest vocabularies by the oldest pupils.
The positive correlation of vocabulary with mental
age is shown in Table 14 for comparison.
40-49 . .
TABLE IS. VOCABXJLARY AND Aon. (Correlation .40)
TABLE 14. VOCABULABY AND MENTAL AQB. (Correlation +.656}
84 INTELLIGENCE OF SCHOOL CHILDREN
I Q and school work. The correlation between I Q
and school work was somewhat higher than between
mental age and school work (.545 as against .44).
While the disagreements were fairly numerous, most
of them could be accounted for by such factors as
health, attendance, degree of application, and attitude
toward work. Often it was the test which disagreed
most with quality of school work that contributed
most to an understanding of the pupil. In general,
however, school work rose and fell with I Q, as is
shown by Tables 15 and 16.
Average I Q
TABLE 15. AVERAGE I Q FOB DIFFERENT SCHOOL MARKS
I Q Average mark
75- 84 63
85- 94 72
125 and over 83
TABLE 16. AVERAGE SCHOOL MARK FOR DIFFERENT I Q's
I Q and teachers' estimates of intelligence. The
teachers were asked to estimate the intelligence of
each pupil on the usual scale of 1, 2, 3, 4, 5. For
DIFFERENCES IN HIGH-SCHOOL PUPILS 85
pupils the ratings were made by at least three teachers.
The ratings for each child were then averaged to secure
a composite rating. The teachers did not confer with
one another in making the ratings nor did they know
the results of the tests. The correlation of the com-
posite ratings with I Q's is shown in Table 17.
1 to 1.74
1.75 to 2. 49
2.5 to 8. 24
3.25 to 3. 99
4 to 4.75
TABLE 17. SHOWING AGREEMENT BETWEEN I Q AND TEACHERS*
RATINGS ON INTELLIGENCE. (Correlation .59)
The correlation is fairly high. It would have been
considerably higher but for the fact that the over-age
children were rated too high, the under-age children
too low. The_tendency of . teachers, is to, base their
estimates of intelligence on the quality of the work,
paying too little attention to age or degree of applica-
tion. The correlation between the teachers' ratings
and the class marks was .70. There were eight pupils
below 95 I Q who received an intelligence rating of
"average." All but two of these were above the me-
dian chronological age of the class.
Although the teachers* ratings were made independ-
86 INTELLIGENCE OF SCHOOL CHILDREN
ently of each other, there proved to be an average inter-
correlation of .677 between the ratings of one teacher
and those of another. This would indicate that all the
teachers based their estimates of intelligence on much
the same thing, namely, quality of school work.
Relation of intelligence to elimination. Of the 107
who entered the Palo Alto High School in 1916-17,
all of whom were tested, there were 27 who did not
reenter the following year. Fourteen of these had
transferred to other schools and IS had left school "to
go to work." The I Q's of the latter group were 79,
83, 85, 87, 90, 9, 97, 97, 101, 105, 106, 115. The boy
with I Q of 115 had left only temporarily on account
of family finances. Ten of the 13 were below the
median I Q for the class (105). The average I Q of
the 14 who transferred to other schools was 110. The
average of the 13 who dropped out was 94. Seven of
the 13 had received marks denoting failure in more
than half their school work. Plainly most of these
pupils did not really "quit school to go to work "; they
went to work out of school because they could not do
the work in school.
Had there been a better understanding of the degree
of mental ability necessary for success in certain stud-
ies fewer eliminations would have resulted. In this
high school, at least, the pupil with I Q below 90 is
practically certain to fail in such studies as algebra
and Latin. For purposes of educational guidance it
will be necessary to establish the lower limits of intel-
DIFFERENCES IN HIGH-SCHOOL PUPILS 87
lectuality necessary for success in the various high-
Other evidence that elimination is selective. In
the average American city not more than 40 per cent
of the pupils who enter the first grade remain to enter
high school, and ordinarily not more than 10 per cent
graduate from the high school. Smaller cities make
somewhat better records, hut it is an exceptional
school system that graduates from the high school as
many as one fifth of its children. In the case of the
318 cities of all sizes studied by Strayer, the central
tendency was for about 37 per cent to enter the first
year of high school, 85 per cent to enter the second
year, 17 per cent the third year, and 14 per cent the
fourth year. 1 The 58 cities studied by Ayres and the
3 studied by Thorndike made a considerably lower
record, particularly in the third and fourth high-school
grades. It is not uncommon for one third to drop out
without completing the work of the first year. Not
all of this elimination is traceable to inferior mental
ability, but that a large part is due to this cause there
is no longer room for doubt.
Van Denburg studied the school records of 1000
representative children who entered the first year of
high school in New York City. That these 1000 pupils
represented a rather highly selected group is shown by
the fact that although only one pupil in twenty-three
1 Strayer, G. D., Age and Grade Census cf Schools and Colleges.
Bull. No. 451, U.S. Bureau of Education, p, 6.
88 INTELLIGENCE OF SCHOOL CHILDREN
in the elementary schools of New York gained special
promotion, one third of those who entered high school
had done so. We have already seen that pupils who
enter high school considerably retarded are almost in-
variably pupils of inferior ability, and that those who
enter under-age are exceptionally bright. Remember-
ing this, it is interesting to note that Van Denburg
found that pupils who enter late are very much less
likely to graduate than those who enter young. The
same result was found for Iowa City over a period of
ten years. 1 Table 18 shows the "graduation expect-
ancy" of pupils who enter at various ages.
TABLE 18. GRADUATION EXPECTANCY OF PUPILS ENTERING HIGH
SCHOOL AT VARIOUS AGES
Even when the late entrant remains to graduate he
normally requires more than four years to do so. For
example, King found that only 18 per cent of those
entering at 16 graduated in four years, and only 9 per
cent of those entering at 17.
Van Denburg's 1000 pupils were rated by their
teachers on ability shortly after they entered upon the
1 King, Irving, The High School Age (Dodd, Mead & Co., 1914),
DIFFERENCES IN HIGH-SCHOOL PUPILS 89
first semester's work. Three grades were used:
"high," "average," and "low." Of those rated
"low" 50 per cent dropped out in one half year or less;
of those rated "average," 50 per cent dropped out
within one and one half years; of those rated "high,"
50 per cent remained for three years or more. The
marks given these pupils at the end of the first term
proved also to have great value as an index of future
elimination. The median expectancy for those secur-
ing various marks was as follows:
Average of 1st term's Time during which 50 per
marks (per cent) cent remained in school
0-49 } year
50- 59 1 "
60- 69 1J years
70- 79 2J "
80-100 4 "
There can be but one conclusion from facts like those
we have just cited: high-school elimination is very selec-
tive. Although there are many individual excegtioS;
the pupils who drop, out are jnjhe main pupils of in-
f erior ability. The high school offers little which can be
done by pupils of much less than average intelligence,
Are high-school standards too high? It would seem
that if the pupils of inferior ability are to be retained
the high school will have to do one of two things:
either (1) lower the standards in the present courses,
or () add other studies which are easier while at the
same time educationally worth while.
90 INTELLIGENCE OF SCHOOL CHILDREN
It may be that we have judged the high school too
exclusively by the difficulty pupils encounter in meet-
ing its standards for graduation. Largely through
the influence of the university, the bars have been
raised until graduation is well beyond the intellectual
endowment of a large proportion of children. Below
90 I Q graduation is by no means likely, and nearly a
third of all children test this low or lower. Proctor
found that 70 per cent of those testing below 95 I Q
failed in more than half of their studies. A nation
falls short of the true ideals of democracy which re-
fuses to furnish suitable training to a third of its chil-
dren merely because their endowment does not enable
them to complete a course of study which will satisfy
the requirements for college entrance.
There was a time when those whose ability would
not carry them through algebra or Latin could turn
with some hope of success to the modern languages or
to science. In proportion as these studies became
established they too raised their requirements. When
the commercial subjects were brought into the high-
school curriculum, these in turn became the dumping-
ground for failures. However, the teachers of com-
mercial subjects were not long in discovering that
there is no demand in stenography or bookkeeping foi
commercial graduates of inferior ability. At present
other lines of vocational training are being introduced
into the high school and the pupils who cannot succeed
in the older subjects are turning to these. Whether
DIFFERENCES IN HIGH-SCHOOL PUPILS 91
the solution will be found there will depend largely on
the variety of courses the high school undertakes to
offer, and on whether it is willing to forego the semi-
collegiate standards in favor of a humbler task.
High schools at present are in a measure "class"
schools. The child of 75 to 85 I Q has an inalienable
right to the kind of training from which he can derive
profit. Since there are so many who cannot master
the usual high-school studies, new lines of work of a
more practical nature will have to be added. Since
there are probably.teivper cent who have not-even the
ability to complete the work preparatory to high school,
the differentiation of courses wilLhave to begrar-in the
sixth or seventh- grade. Instead of being undemo-
cratic, as some have argued, such differentiation of
courses and enlargement of opportunities for vocational
training of the humbler sort is a necessary corollary, of
the truly democratic ideal.
THE MENTAL-AGE STANDARD FOR GRADING *
THE I Q does not itself tell us in what grade a pupil
belongs. A child testing at 75 I Q and another testing
at 125 may be equally ready for work of fourth-grade
difficulty, provided the chronological age of the former 1
is thirteen and that of the latter eight. Each would
thus have a mental age of approximately ten years.
The basis of grading is therefore mental age rather
tharul (J. The latter is merely an index of brightness;
It is extremely significant, because it enables us i&
forecast a child's later mental development, but grade
of work which a pupil can do at any given time de-
pends rather upon the absolute mental level.
There is a slight correction to add to this statement.
To a certain extent I Q differences do affect the qual-
ity of school work which a given mental age may do.
In the illustration given above, it is altogether likely
that the eight-year child of 125 I Q will do somewhat
better work in the fourth grade than the thirteen-year
child of 75 I Q, even though they have the same men-
tal age. The greater intellectual spontaneity of the
young bright child somewhat outweighs the advantage
which the older but mentally inferior child has in age
and school training.
1 Written with the assistance of Isabel Preston.
MENTAL-AGE STANDARD FOR GRADING 93
Normal mental age for the different grades. The
child is expected to start to school between the ages of
six and seven years. Although many start later and
some younger, the average entrance age in most parts
of the United States is not far from six and a half.
Reckoning on this basis the standard mental age for
the different grades would be as follows :
Grade Standard mental age
1 6-6 to 7-5 or approximately 7 years
H 7-6 to 8-5' " 8 "
HI 8-6 to 9-5' " 9 "
IV 9-6 to 10-5' " 10 "
V 10-6 to 11-5' " 11 "
VI 11-6 to 12-5' " 12 "
VH 12-6 to 13-5' " 13
VEI 13-6 to 14-5" " 14 "
High School I. ... 14-6 to 15-5" " 15 "
Children who are in grades corresponding to these
standards are in the large majority of cases found
doing work of average quality. If the mental age is
much above or below the norms just indicated the
school work is usually correspondingly superior or
Table 19 shows the per cent of children rated as
superior, average, or inferior who are in the grade cor-
responding to mental age (19S6 cases).
It is seen that the mental age norms we have given
fit the difficulty of work in the different grades fairly
closely. There is a slight tendency, however, for chil-
94 INTELLIGENCE OF SCHOOL CHILDEEN
Quality of work
6-6 to 7-5
7-6 to 8-n5
8-6 to 9-5
0-6 to 10-5
10-6 to 11-5
11-6 to 12-5
12-6 to 13-5
13-6 to 14-5
14-6 to 15-5
TABLE 19. SHOWING QUALITY OF SCHOOL WORK DONE BY CHILDREN
WHO ARE IN A GRADE CORRESPONDING TO MENTAL AGE
dren of the mental age 6-6 to 7-5 to do better than
average work in grade I, and for those of mental age
13-6 to 14-5 to do below average work in grade VIII.
This is what should be expected, since the average
mental and chronological ages for grade I are a little
below seven years, and those of grade VIII a little
above fourteen years. In the first year of high school
the child of standard mental age finds it still more
difficult to do average work. The median mental ages
actually found in the eight grades and the first year of
high school are as follows:
H 8 L
MENTAL-AGE STANDARD FOR GRADING 95
So far, we have shown that the child of standard
mental age for a given grade tends to do average work
in that grade. It remains to show that if the mental
age is above or below the standard, the school work
tends to be superior or inferior to the average.
Of the 1936 children appearing in the above table,
120 were two or more years above the grade normal to
their mental age. This is 6.8 per cent of the entire
number. Of thel&O not one was rated as doing superior
work, and only 19 as doing average work. The remain-
ing 101 were rated as doing work of inferior or very
inferior quality. Of the 1936 there were 234 who were
located in a grade two or more years below the stand-
ard for their mental age. Of these, 52 per cent were
rated above average in school work, 33 per cent aver-
age, and 15 per cent below average.
Summarizing, we can say that while children located
in a grade two years above mental age are rarely able
to do average work, there are somewhat more in a
grade two years below mental age whose school work
is not satisfactory. The child with mental age more
than equal to his work may yet fail because of illness,
lack of application, or for any of a number of reasons.
On the other hand, exceptional industry can rarely
make good the disadvantage which a child suffers
whose mental age is two or three years below the
Sources of error in judging school success. The
agreement of school performance with mental age
96 INTELLIGENCE OF SCHOOL CHILDBEN
standards would doubtless have been closer if all the
teachers who rated their children had been infallible
udges of the quality of school work, if the work of a
given grade had always been of the same difficulty,
and if they all had taken the terms " average/ 5 " supe-
rior," and " inferior " in exactly the same sense. All
of these sources of error are serious, especially the last.
As we have pointed out so many times, each teacher
tends to take as her rating standard the average work
actually being done in her class. If her class has a
disproportionate number of dull pupils she tends to
rate too high; the reverse, if her class as a whole is
Ratings on school work are also likely to be influ-
enced by the personal traits of the individual children.
Traits which tend to cause over-rating are vivacity,
responsiveness, talkativeness, self-confidence, good
looks, neatness, application, and conscientiousness.
The child who is vivacious and self-confident, but
parrot-like and superficial, is almost sure to be over-
rated; the stolid-appearing or quiet and timid child,
to be under-rated. The child who does his work
neatly and conscientiously is likely to be rated more
leniently than the child who is slovenly, careless, or
disobedient. The child whose hearing or speech is
defective is also at a disadvantage in such compara-
Errors of this kind, however, are not sufficient to
account for the fact that only forty to sixty per cent
MENTAL-AGE STANDARD FOR GRADING 97
of school pupils are located in the grade corresponding
to mental age. Perhaps an even more frequent cause
of incorrect grading is the tendency of teachers to
promote children by age, resulting in over-promotion
of the dull and under-promotion of the bright. The
teacher does not ordinarily realize how far the dull
over-age child has been promoted beyond the grade
where he could do average work. She is still farther
from knowing that the typical under-age bright child
would in a majority of cases continue to do satisfac-
tory work if promoted one or two grades.
However, there are occasional discrepancies be-
tween mental age and school performance which can-
not be traced either to errors in rating or to mechani-
cal methods of promotion. The quality of a child's
school work depends in part upon other factors than
intelligence, among which are health, regularity of
attendance, degree of application, attitude toward
teacher, emotional stability, amount of encouragement
at home, etc. The effect of most of these extraneous
factors is to make school performance less satisfactory
than the mental age would lead us to expect.
Discrepancies between mental age and school per-
formance. For several years, in connection with Binet
tests made by many Stanford University students, we
have investigated those cases in which a marked dis-
agreement was found between mental age and school
performance. The findings would fill a long and in-
teresting chapter, but the results of a single series of
98 INTELLIGENCE OF SCHOOL CHILDREN
tests will acquaint the reader with the common types
of cases. We will select for this purpose the investi-
gation of Miss Preston, who made a study of the dis-
agreements found in tests of 238 pupils in the eight
grades of the Santa Clara, California, grammar school.
The pupils tested constituted about a third of those
enrolled in the school, and were selected so as to be as
nearly as possible representative. Most of them had
also been given the Trabue B and C Completion Tests
and the Army mental test. In addition each child
was rated by the class teacher on each of the following:
social status, school work, intelligence, dependability,
and social adaptability. Miss Preston had been for
ten years principal of the school in which the tests
were made, and had known all the children personally
from the time they first entered. Her acquaintance
with parents and home conditions was also of great
It was found that in the great majority of cases the
result of the Stanford-Binet test agreed remarkably
well with the child's school work, particularly when
the quality of work for a period of years was made the
basis of the comparison. The 238 tests yielded only
"84 discrepancies worthy of note, and many of these
were not large. In 29 of the 34 cases the quality
of school work as rated by the teacher was poorer
than the mental age would seem to warrant, and
in only 5 cases better. Where discrepancy of the
latter kind occurred it was ordinarily due either to ex-
MENTAL-AGE STANDARD FOR GRADING 99
ceptional application on the part of the child or to the
effect of vivacity, responsiveness, or other favorable
personal traits in influencing the teacher's judgment.
On the other hand, discrepancy in the direction of in-
ferior work resulted from a variety of causes, including
timidity, lack of self-confidence, physical defects, lack
of application, emotional instability, psychopathic
heredity, home " spoiling," love affairs, etc.
In the following pages we present Miss Preston's brief
description of the salient features of typical cases of
discrepancy. The ratings given by the teacher were in
this experiment based on a scale of seven, 1 as follows:
1. "Very superior";
3. "Above average";
5. "Below average";
7. "Very inferior."
Showing effect of unusual application.
Ernest. Age 15-0, mental age 12-3, I Q 82, eighth
grade, quality of work 5. Portuguese, social status 5.
Teacher's ratings: intelligence 5, social adaptability 8, de-
Discrepancy: The mental age is a year and a half below
that normal for the eighth grade, but the work is passing
(though below average).
1 It will be necessary to bear in mind that the numbers designat-
ing ratings do not correspond to those quoted in other chapters of
this book, most of which were based on a five-fold instead of a seven-
100 INTELLIGENCE OP SCHOOL CHILDREN
Ernest's teachers agree that the .test has rated his intellec-
tual ability correctly. It happens, however, that his most
characteristic trait is one which escapes an intelligence scale.
Ernest is an erect little fellow, with a straightforward look,
who works with all his might at anything he attempts to do.
No other pupil in the school equals him in application. He
often reaches school ten minutes after seven o'clock in order
to study his history lesson until nine. First he reads and
re-reads his lesson in an attempt to get the meaning. Then
he writes it. After that he says it aloud over and over.
When the other children begin to arrive he hauls one of them
in to hear him recite it and to have him explain what some of
the words mean. In class, Ernest is a living question mark.
"What does it mean by those words?" "Does it mean
this?" He is oblivious to the teacher's impatience and to
the amusement caused among his fellow pupils. There is
no escape from his questions. Even as the line files in or
out his teacher gets a "What does this mean?" as he
marches by. When Ernest does a thing he does it thor-
oughly, in school or out. He is captain of the baseball team
and docs a vigorous job of it.
Showing effect of child's personality on the teach-
Jennie. Age 12, mental age 10-8, I Q 89, sixth grade,
quality of work . American, social status 3. Teacher's
ratings: intelligence 2, social adaptability 2, dependability 2.
Discrepancy: "Superior" work in a grade which is more
than a year above her mental age.
Jennie attracts attention by her smiling, vivacious face
and sparkling eyes. She is alert, quick in movement, but
without self-consciousness. In conversation she is respon-
sive, eager and reflects your every expression. In class her
eye never leaves the teacher's face and she follows every ex-
planation with intent eagerness. All of this naturally influ-
MENTAL-AGE STANDABD FOR GRADING 101
ences the teacher's estimate of her intelligence and school
Donald. Age 12-0, mental age 16-8, I Q 189, sixth
grade, quality of work 2. American, social status 3.
Teacher's ratings: intelligence 2, social adaptability 2, de-
Discrepancy: The mental age would indicate ability to do
high school work, but Donald is only in the sixth grade.
However, his work hi this grade is "superior" and it is prob-
able that he would be able to do satisfactory work in a higher
Donald is chiefly of interest hi comparison with Jennie,
described above. The two are of almost exactly the same
age and are both doing "superior" work hi the sixth grade.
Jennie, however, is barely average-normal in intelligence,
while Donald tests at 139. This difference is confirmed by
the Trabue test and also by the Army test.
In personality Donald presents a striking contrast to Jen-
nie. Her responsiveness and vivacity are fully matched
by his apparent stolidity and shyness. Donald talks only
in monosyllables. He has been so thoroughly suppressed
at home by a severe father that he is shrinking and timid.
When successful in drawing him out one finds a highly sensi-
tive nature of rare sweetness and poetic feeling. But the
least stir sends him shrinking back into his shell with a hurt
air and a suspicious glance. He has no self-confidence,
never expresses his feelings, and avoids doing anything that
could possibly attract attention.
Claire. Age 9-10, mental age 12-7, I Q 128, fourth
grade, quality of work 2. American, social status 4.
Teacher's ratings: intelligence 2, social adaptability 2, de-
Discrepancy: Mental age two years above her grade.
However, her school work is superior and she could proba-
bly do the work in the next higher grade. The fact that
108 INTELLIGENCE OP SCHOOL CHILDREN
she has had one extra promotion agrees with her high in-
Claire is slow in her movements and slow in finishing
assigned tasks. She is diffident, hesitating in speech, and
waits for approval. Her teacher seldom realizes, until re-
view time, how thoroughly Claire gets her work.
Showing effect of timidity and lack of self-confi-
Clifford. Age 8-5, mental age 8-6, I Q 101, third
grade, quality of work 5. American, social status 4.
Teacher's ratings: intelligence 4, social adaptability 4, de-
Discrepancy: In grade corresponding to mental age, but
his work until last year barely passing.
Clifford has no self-confidence. His .mother speaks in
his presence of his stupidr and compares him disparag-
ingly with his bright older brother. Hard to get Him to try,
but his work has recently shown improvement.
Louise. Age 9-1, mental age 10-6, I Q 115, fourth
grade, quality of work 6. American, social status 3.
Teacher's ratings: intelligence 4, social adaptability 4, de-
Discrepancy: School work "inferior" although mental age
would indicate ability for the fourth grade.
Louise is timid and easily worn out by excitement. Likely
to appear bewildered when placed in a group. Is domi-
nated by an older sister whom she worships, but who has
reached an irritable stage in her development. Louise can-
not please her in any way, although her endeavors are con-
Showing effect of mental inertia.
Leonard. Age 13-6, mental age 13-10, 1 Q 103, seventh
grade, quality of work 6. American, social status 5.
MENTAL-AGE STANDABD FOB GRADING 103
Teacher's ratings: intelligence 5, social adaptability 5, de-
Discrepancy: Both chronological age and mental age
normal for grade, but school work has always been decidedly
Leonard's father, now dead, was a shiftless drunkard.
The mother, ostensibly a nurse, leads an immoral life.
Several cousins are feeble-minded. Leonard's smiling good
nature and constitutional indolence are proverbial among
his teachers. One wonders whether he ever did anything
he was not compelled to do. In school he sits smiling pleas-
antly at others or staring off into space, dreaming. When
prodded by the teacher he opens his book and stares into
it vacantly. Perhaps the book is upside down. Occasion-
ally he wakes up and gives a clear, fluent account of some-
thing he has read or seen, but he soon lapses again into his
customary state of oblivion.
Showing effects of emotional instability or nervous
Olivia. Age l$-6, mental age 13-2, I Q 105, seventh
grade, quality of work 6. Portuguese, social status 4.
Teacher's ratings: intelligence 5, social, adaptability 4, de-
Discrepancy: Up to grade-age mentally and chronologi-
cally, Olivia has been promoted on trial from almost every
Of Portuguese parents whose heads have been turned by
prosperity. The mother says in her presence that Olivia
has inherited her own nervousness and inability to do arith-
metic. Needless to say, Olivia is nervous and cannot do
arithmetic. She flounces around at her lessons, adds a bit,
jiggles her desk, drops her books, picks them up, caresses
her curls, etc. When in trouble pretends to be about to
faint, but quickly recovers if threatened with punishment.
104 INTELLIGENCE OF SCHOOL CHILDREN
Emotional instability fully explains the discrepancy be-
tween intelligence and school success.
Joseph. Age 13-10, mental age 15-9, I Q 114, eighth
grade, quality of school work 6. American, social status 5.
Teacher's ratings: intelligence 4, social adaptability 6, de-
Discrepancy: Mental age above average for grade, but
school work very unsatisfactory.
Joseph has two sisters who are feeble-minded and blind.
Two of his three brothers, also feeble-minded, are dead. The
third brother is a "movie" star of national fame. Joseph's
mother is a kindly faced woman who has been deserted by
her worthless husband, and supports herself by taking hi
washing. Joseph himself is a bookworm, reading every-
thing he can lay his hand on from Sunday-School books to
encyclopaedias. His mind is an exhaustless reservoir of
unrelated facts. Psychopathic symptoms. Suffers at times
from the idea of persecution, at which times he refuses to do
any school work or even to talk.
Effect of home " spoiling."
Gordon. Age 5-7, mental age 6-6, 1 Q 116, first grade,
quality of work 5. American, social status 3. Teacher's
ratings: intelligence 5, dependability 5.
Discrepancy: School work "inferior," although in grade
corresponding to mental age.
Gordon is the son of a minister and badly spoiled from
petting and humoring. Attitude of condescension toward
school work. Attention poor; easily fatigued; bad sex habits.
Bernard. Age 7-1, mental age 7-B, I Q 108, first grade,
quality of work 5 to 6. Portuguese, social status 5. Teach-
er's ratings: intelligence 4, social adaptability 4, dependa-
Discrepancy: School work below average, although men-
tal age is a half year above normal for grade.
MENTAL-AGE STANDARD FOR GRADING 105
Bernard is a handsome child; the other five in the same
family are very homely. Has always been petted and al-
lowed to have his own way.
Showing influence of physical defects.
Roy. Age 15, mental age 14-8, 1 Q 98, seventh grade,
^aality of work 6. American, social status 4. Teacher's
ratings: intelligence 6, social adaptability 5, dependability 5.
Discrepancy: Mental age a little above the average for his
grade, but school work "inferior."
Thin, anaemic, and sickly-looking. Almost hydrocepha-
lic in appearance, with protruding eyes and open mouth.
Very deaf and resents it. Fails to hear much of what is
said during recitation, but will not admit it. At home has
been alternately scolded and petted by a foolish mother,
with the result that he has irritable and stubborn spells.
Madeline. Age 7-10 (?), mental age 6-2, 1 Q 79 (?), first
grade, quality of work 7. Portuguese, social status 5.
Teacher's ratings: intelligence 6, social adaptability 6, de-
Discrepancy: Although there is a question about Made-
line's correct age, her mental age of 6 plus should enable her
to do at least fair work in the first grade. She is making al-
most no progress.
Has suffered for years from chorea. Attends school until
her movements become too uncontrolled and violent, stays
at home for a few weeks, then returns to school. After a
severe attack all she has learned in school seems to leave
Love affairs and day-dreaming.
Elmer. Age 14-2, mental age 14-3, I Q 101, seventh
grade, quality of work 6. American, social status 4.
Teacher's ratings: intelligence 5, social adaptability 3, de-
106 INTELLIGENCE OF SCHOOL CHILDREN
Discrepancy: Failing in work, but with mental age
slightly above his school grade.
The discrepancy in Elmer's case was only temporary and
was caused by a particularly severe case of "puppy love."
The girl moved away, love's young dream was broken, and
Elmer's work came back to normal.
Aldrich. Age 7-5, mental age 8-6, I Q 115, second
grade, quality of work 6. American, social status 4.
Teacher's ratings: intelligence 3, social adaptability 3, de-
Discrepancy: Mental age average for grade but school
A dreamer and not interested in school work. Poor
foundation in first grade. Teacher's estimate of intelli-
gence agrees with the I Q. Unusual vocabulary and in-
Summary. It appears that lack of self-confidence,
personal traits which tend to cause over-rating or un-
der-rating, mental inertia, physical defects, emotional
instability, and psychopathic heredity are the most
common causes of discrepancy between mental age
and quality of school work. Unfavorable emotional
attitude toward the teacher, the effects of which we
have seen in other cases, did not appear in this series.
Of the 34 pupils for whom a discrepancy was found,
24 were boys, although as many girls as boys were
tested. This would indicate either that teachers
oftener misunderstood boys and of tener under-rate
their school work, or that the school performance of
boys is more easily influenced by physical or emo-
tional defects than that of girls.
BIENTAIrAGE STANDARD FOR GRADING 107
It is also interesting to note that although the tests
were almost equally divided between children of Ameri-
can and foreign parentage (chiefly Portuguese), the lat-
ter account for only 11 of the 34 discrepancies. It ap-
pears, therefore, that the fact of foreign parentage does
not greatly limit the usefulness of the Stanf ord-Binet
scale as a measure of a child's educability . Several other
Stanford students have made studies similar to that of
Miss Preston's, involving in all nearly two thousand
Binet tests. The data show convincingly that in the
large majority of cases mental age offers a fairly accu-
rate index as to the grade in which the child is fitted to
do work of average quality. The index misses the
mark to the extent of one grade in something like six to
eight per cent of cases, to the extent of two grades in
not more than one or two per cent of cases. In ninety
cases out of a hundred it is accurate enough for all
practical purposes. Even in those instances where it
would be misleading, if taken as the sole criterion,
the Binet test offers the best available starting point
for reaching an understanding of the child's case.
For example, J. F. had been for months doing very
inferior work in the first year of high school. The
teachers and principal were at a complete loss to un-
derstand the case. Various remedies were tried, but
without effect. The boy claimed that he was making
every effort to do the work. Finally he was given a
Binet test and was found to have a mental age well
above that necessary for successful work in the ninth
108 INTELLIGENCE OF SCHOOL CHILDREN
grade. The principal then called the boy to his office,
explained to him what the test had revealed regarding
his ability, and suggested that it was time for him to
" quit fooling " and get down to earnest work. The
result was an immediate and surprising improvement
in his class marks.
Sometimes the fault lies not so much in lack of ap-
plication as in failing self-confidence. S. W., a boy of
twelve years, had developed a sense of mental inferi-
ority. His school work had gradually deteriorated un-
til he was on the point of failing. Although it is ordi-
narily not permissible to give a child his Binet test
score, the principal wisely decided to do so in this case.
The boy was so encouraged by the information that he
went to work with a new spirit and soon ranked above
the average in his class.
Whether the child is working exactly up to his ca-
pacity, or above or below it, the mental test is equally
necessary. Ernest, the first of Miss Preston's cases, is
doing fair work, although considerably below the nor-
mal mental age for his grade. Unless this is known
Ernest's efforts cannot be correctly appraised. La such
cases as Roy or Madeline the teacher's attention is
directed by the test to the possible influence of physi-
cal defect upon school work. A discrepancy like that
shown by Jennie and Donajd calls attention to the
danger of over-rating the vivacious or under-rating
the diffident child.
The case of Margaret. The case of Margaret, re-
MENTAL-AGE STANDARD FOE GRADING 109
ported by Strong, offers a classical example of the use-
fulness of mental tests in discovering the causes of
poor school work. 1
Margaret had just failed of promotion from the low-
fourth grade. She was eleven years old, and tested at
eleven by the Binet scale. With average normal abil-
ity, according to the test, her school work was never-
theless described by teachers as " hopeless." Her work
in arithmetic and geography was especially poor.
Prom January until May a small amount of special
instruction was given her by one of Dr. Strong's stu-
dents. Although the special instruction in arithmetic
extended over only five months, and amounted to a
total of only a few hours, Margaret's advancement
was from third-grade work to fifth-grade work, as
shown by the Courtis tests.
The trouble seems to have been largely one of emo-
tional attitude. When the special instruction began
" she was afraid of everything; she could do very lit-
tle, she knew nothing positively. She held her eyes
down, carried herself shrinkingly, was a typical *f raid-
cat.' We started with a thoroughly disheartened
child, whose enthusiasm and hope were about dead,
and who was being taught many things in school with-
out knowing facts and principles which should have
preceded these things. We taught her the funda-
mentals of arithmetic, thus filling in all the gaps in her
1 "The Development of Proper Attitudes toward School Work";
in School and Society, December 25, 1915, vol. n, pp. 926-34.
110 INTELLIGENCE OF SCHOOL CHILDREN
knowledge of that subject up to the work of her class.
In doing so we allowed her to see her learning curves.
The unmistakable objective fact that she was learning
made her realize that she could learn, aroused her in-
terest, gave her fresh enthusiasm, and presently there
resulted a transformed child. As we have seen, the
transformation affected not only arithmetic, but all
her studies, her carriage and walk, her social attitude
toward others, her entire character. From being
hopelessly at the bottom of her class she now has a
settled determination to lead that class. From every
indication it appears that the actually brighter chil-
dren will have to work to keep ahead of Margaret."
MJfiJN JL'AL TESTS OP SCHOOL LAGGARDS
SINCE the publication of Ayres's book, Laggards in
Our Schools, 1 numerous statistical studies have been
made showing the large proportion of children who
are one, two, three, or more years retarded, and ana-
lyzing the factors which are responsible for the condi-
tion. It has become a matter of common information
that more than ten per cent of the cost of tuition is for
repeated instruction, that about a fourth of the pupils
leave school with not more than a sixth-grade educa-
tion, and that the ranks of the vocationally incompe-
tent are recruited largely from children who in school
were over-age for their grade. Yet the problem re-
mains. The number of school laggards has decreased
but little, and their needs are almost as little provided
for as before the campaign in their behalf began.
The extent of the problem will be apparent from an
examination of typical statistical findings.
The number of over-age children. Professor
Strayer by a uniform method secured data on the
amount of retardation in 132 cities having a popula-
tion of more than 5,000, and in 186 cities having a
1 Ayres, Leonard P., Laggards in Our Schools. 1909. Russeft
Sage Foundation; pp. 236.
lia INTELLIGENCE OF SCHOOL CHILDREN
population less than 25,000. 1 His most important
results are embodied in Table 20, which show for boys
and girls separately the average per cent found re-
tarded or accelerated by various amounts. 2
TABLE 20. RETARDATION IN 318 CITIES. (Strayer)
Table 20 shows that approximately one child in three
is retarded, and only one in twenty-five accelerated. More
than five per cent are retarded three years or more, and
nearly fifteen per cent two years or more.
The actual amount of retardation is even greater
than the figures indicate, because of the liberal basis
on which retardation was computed. The standard
adopted by Professor Strayer was that used by Ayres,
By this standard a child is considered as 'making nor-
mal progress if in the first grade and not yet eight
1 Strayer, G. D., Age and Grade Census of Schools and Colleges.
Bull. 451, U.S. Bureau of Education, 1911; pp. 144.
* Professor Strayer has omitted fractional per cents, so that the
Combined totals do not quite equal 100 per cent.
MENTAL TESTS OF SCHOOL LAGGABDS 118
years old, if in the second grade and not yet nine years
old, and so on* Although this is the standard on
which retardation is nearly always computed, it allows
a margin of a half year all along the line. Actually, if
the child begins school by the age of six and one half
years and makes normal progress, he will be in the
second grade by the age of seven and one half, in the
third by the age of eight and one half, etc. Counting
retardation on this basis, the figures for Salt Lake
City, in May, 1915, were as follows:
Over-age 1 year 26.7 per cent
Over-age 2 years 11. 2 per cent
Over-age 3 years 3. 7 per cent
Over-age 4 years or more 1.2 per cent
Total over-age 43. percent
That the retardation evil is not confined to large
cities is shown by Strayer's figures for cities under
25,000; also fey Berry's, Lurton's, and Morton's statis-
tics for small towns in Michigan, Minnesota, and Ne-
braska. For 55 cities and villages in Minnesota, the
number of children retarded one year or more (Ayres's
basis of reckoning) was 30.9 per cent, and in 41 graded
schools of the same State the number was 83.9 per
cent. 1 The figures for 96 cities and towns of Nebraska, 2
taken together, are as follows:
i Lurton, F. E., "Retardation Statistics from the Smaller Minne-
sota Towns"; Psychological Clinic, 1911. m
* Morton, W. H. S., "Retardation in Nebraska"; Psychological
Clinic, December, 1912, and January 19, 1913.
114 INTELLIGENCE OF SCHOOL CHILDREN
Total pupils 25,449
Retarded 1 year 16 .3 per cent
Retarded 2 years 7.6 per cent
Retarded 3 years 3.3 per cent
Retarded 4 years or more 1.4 per cent
Total retarded 28.6 per cent
In 227 cities and towns of Michigan, Berry found a
total of 24 per cent retarded, and 3.5 per cent retarded
three years or more. 1
The amount of retardation in rural schools seems to
be even greater than in cities and towns. Gaylor
found 53.6 per cent of the children in 139 rural schools
of an Illinois county at least one year retarded, and
28.4 per cent more than one year retarded. 2
In 11 small cities in Illinois the number retarded
two years or more was 20.7 per cent, as compared with
28.4 per cent in the rural schools.
Phelps found 28 per cent of 13,626 rural-school chil-
dren in California retarded on the liberal standard
used by Ayres. The number retarded three or more
years was 2.5 per cent.
The supposed causes of retardation. Retardation
cannot be properly dealt with until its causes are un-
derstood. The causes emphasized by Ayres and the
majority of other investigators are physical defects,
1 Berry, Charles Scott, "A Study of Retardation, Acceleration,
Elimination, and Repetition in the Public Elementary Schools of
225 Towns and Cities of Michigan"; Seventy-ninth Annual Report of
Superintendent of Public Instruction (Michigan), 1915-16.
* Gaylor, G. W., "Retardation and Elimination in Graded and
Rural Schools"; Psychological Clinic, 1010, pp. 40-45.
MENTAL TESTS OF SCHOOL LAGGABDS 115
irregular attendance, late entrance, too high a stand-
ard, and lack of flexibility in methods of promotion.
The usual view is expressed in the following conclu-
sions which Dr. Gulick drew from the investigation of
Ayres: (1) "That the most important causes .of re-
tardation can be removed "; (&) " that relatively few
children are so defective as to prevent success in
school or life." These assumptions are contradicted
by the findings of all who have investigated the sub-
ject by the use of mental tests.
It is interesting to examine the causes which are
most often mentioned by teachers and superintend-
ents. In the case of 108 laggards in the Salt Lake City
schools, the causes most often named were the follow-
ing, given here in order of frequency of mention by
teachers: poor home condition, physical defects,
transferring from another school, retarded mental de-
velopment, difficulty with the English language, lack
of application, irregular attendance, laziness, late en-
trance, and delinquency.
Mental tests of these same 108 children showed an
average mental retardation of three years. The large
majority, indeed, were feeble-minded. Feeble-minded
children do often come from poor homes, since often
the parents of feeble-minded children are themselves
feeble-minded. For the same reason, feeble-minded
pupils shift frequently from one locality to another
and attend irregularly. Because such children are
feeble-minded, they enter late, show little applica-
116 INTELLIGENCE OF SCHOOL CHILDREN
tion in their school work, and tend to become delin-
Many similar questionnaire studies have been made
and their findings are always essentially the same.
All kinds of supposed causes of retardation are empha-
sized except the one important cause inferior men-
tal ability. Many teachers seem to hold views some-
what as follows: (1) All children are either normal or
feeble-minded; (2) those who are normal (i.e., not
feeble-minded) should make standard school progress;
(3) those who are feeble-minded will bear readily rec-
ognizable ear-marks of their deficiency and will be un-
able to learn anything. It is not generally understood
that many feeble-minded children present a normal
appearance; still less that some ten per cent of school
children of perfectly normal appearance have a grade
of intelligence which is about halfway between that of
the moron and the average normal child.
The real cause of retardation. Of Dickson's first-
grade pupils who were eight years old or older, 68 per
cent were below 80 I Q. Of Hubbard's fifth-grade
pupils who were thirteen years old or older (i.e., two or
more years retarded) 64 per cent were below 80 I Q.
Of 50 over-age children tested by Williams in three
California cities, 50 per cent had an I Q below 80, and
38 per cent were below 75. Of 174 over-age children
tested in the schools of " X " County, California, 61
were below 70 1 Q, 106 below 81 and 153 below 90. In
the case of the 1000 unselected children on whom the
MENTAL TESTS OF SCHOOL LAGGARDS 117
Stanford Revision was based, 97 were three or more
years over-age on the Ayres standard. Of these,
78 per cent were below 80 I Q. Conversely, nearly
all of those who were below 80 I Q were one or
more years over-age for the grade in which they were
Dozens of such studies, larger or smaller, could be
quoted. It is unnecessary, for all show the same
thing: namely, that the over-age child is usually a dull
child. Any one who desires additional proof need only
test a large number of unselected children of a given
chronological age, say twelve years, and note the
school progress which those of various mental ages
have made. Tables 21 and 22 show this for unselected
children of eleven and twelve years.
TABLE 21. GRADE LOCATION OF 263 H-YEAR-OLDS BY STANFOBD-
BINET MENTAL AGE
These children were all eleven years old, chronologically. Correlation is .81
118 INTELLIGENCE OF SCHOOL GHILDBEN
TABLE 22. GRADE LOCATION OF 257 12-YsAB-OLDs BY STANFORD-
BINET MENTAL AGE
These children were all twelve years old, chronologically. Correlation is .856
In the above tables, mental age 6 means 5-6 to 6-5,
7 means 6-6 to 7-5, and so on. The tables show that
intelligence is the chief factor determining the rate of
a child's progress. It also shows that the test result
gives a fairly reliable indication of the school grade in
which a child of a given chronological age will be able
to do the work. The correlation between mental age
and grade is .81 for the eleven-year-olds, and .855 for
The 257 pupils of the twelve-year group belong in
the sixth grade by chronological age. However, 47 are
in the fourth grade or below; i.e., two or more years
retarded. All but three of these are mentally below
twelve years, and all but 9 mentally below eleven. Of
MENTAL TESTS OF SCHOOL LAGGARDS 119
the 17 twelve-year-olds in the seventh or eighth grade
(two years accelerated) all but one are mentally four-
teen or above. Similarly for the eleven-year olds.
Those who are accelerated show a high mental age,
those who are retarded test low.
Feeble-minded school children. We have seen that
the large majority of over-age children are below aver-
age intelligence. Such children may be classified as
dull, border-line, or feeble-minded. While the feeble-
minded group is much the smallest of the three, it
gives rise to the most difficult educational and social
problems. What these problems are can best be il-
lustrated by the results of a typical survey of feeble-
minded children in a small school system. That of
" X " County, California, will serve the purpose. 1
"X" County enrolls somewhat more than five
thousand pupils in its public schools. Approximately
twenty per cent of these attend rural schools having
less than three teachers. The other eighty per cent
are divided not very unequally among a half-dozen
small cities. It was not possible to test all these chil-
dren, nor was it necessary to do so in order to ascer-
tain the approximate number of feeble-minded. The
plan adopted was to test the suspected cases in all the
rural schools of the county and in " Y" city, and at
the same time to obtain data from all of the other
1 Terman, Lewis M., Dickson, Virgil, and Howard, Lowery.
The results are published in a bulletin of the California State
Board of Charities and Corrections entitled Surveys in Mental
Deviation, 1918, pp. 19-45.
130 ESTTELUGENCE OF SCHOOL CHIIDBEN
cities of the county of such a kind as would indicate
whether the proportion of mental deficiency in those
cities differed greatly from that found in the schools
where mental tests were given.
The first step was to obtain from the teachers in-
formation which would make possible the location of
suspected cases. At the request of the county each
teacher furnished the following data for each pupil en-
rolled in her class: name, age, grade, years in school,
birthplace and occupation of parents, and ratings of the
child for intelligence and quality of school work as very
superior, superior, average, inferior, and very inferior.
The inf onnation thus secured made it possible to
eliminate eighty or eighty-five per cent of the children
from consideration, because of their obvious normal-
ity. In most classrooms it was necessary to test only
ten to fifteen per cent of the children in order to avoid
the risk of missing any defectives. In certain rooms,
however, more were tested. The rule followed was to
test every child who was rated by the teacher as seri-
ously below average in either school work or intelli-
gence, and to test all who were seriously over-age foi
their grade, whatever the teacher's rating. Of the
1464 pupils enrolled in the rural schools and " Y" City,
174 (12 per cent) were tested. The resulting I Q's
were as follows:
MENTAL TESTS OF SCHOOL LAGGARDS
The majority of cases falling below 70 may be con-
sidered feeble-minded. The range 70-79 is composed
largely of border-zone cases. Those between 80 and
39 are practically always normal, but dull; those be-
tween 90 and 109 may be called average-normal.
In the classification of the 174 suspects only those
were placed below the border-zone group who were
rather definitely feeble-minded. Correspondingly,
those who were above suspicion of feeble-mindedness
were placed above the border-zone group. On this
basis 62 children, or 4.24 per cent of the enrollment of
1464, were classified as feeble-minded, and 29 (1.98
per cent) as border-zone cases.
Grade progress of the feeble-minded. The school
progress which the 62 feeble-minded children of " X "
County were making is shown in Table 23. In
TABLE 23. AGE-GRADE LOCATION OP 62 FEEBLE-MINDED CHILDREN
122 INTELLIGENCE OF SCHOOL CHILDREN
table, as before, age 6 means 5-6 to 6-5; age 7 means
6-6 to 7-5, etc.
From the facts set forth in the above table one could
safely infer, even without the aid of mental tests, that
a majority of these children are very inferior. More-
over, for two reasons the age-grade distribution of the
TABLE 24. GRADE DISTRIBUTION OF 62 FEEBLE-MINDED CHILDREN
BT MENTAL AGE
children represents their mental status too favorably:
(1) The younger feeble-minded have not yet had time
to fall below grade. The feeble-minded of ages 6 and
7, for example, are represented in the table as being up
to grade. (2) The majority of the feeble-minded are in
reality above the grade where they can do satisfactory
work. This is seen in Table 24, which shows that
MENTAL TESTS OF SCHOOL LAGGARDS 123
these children who appear to be so badly retarded
are, on the basis of mental age, greatly accelerated.
While the average retardation on the basis of chron-
ological age is 2.5 years, the average acceleration on
the basis of mental age is 2.2 years. 1
Some exceptionally difficult classes. The follow-
ing schools will give an idea of the problems which
face some of the teachers of " X " County:
Rural School A. Pupils enrolled, 41. Of these, 18 were
so seriously over-age and were rated so low by the teacher as
to be classed as suspects. Of the 18 tested, 13 were feeble-
minded and 3 of borderline intelligence. One family fur-
nished 6 of the feeble-minded, another 4. The school enrolls
one pupil in the first grade who is 10 years old and has been
in that grade for four years. Two other pupils have com-
pleted only two grades in the six years they have attended.
They are now at the age of almost 18 years in the low third
grade and are doing unsatisfactory work there. Another
who is 16 years old and in the seventh grade has only nine-
year intelligence. His intelligence is barely equal to fourth-
Rural School B. Eighty-four pupils, 3 teachers. Of the
12 children tested as suspects, 4 were feeble-minded, 5 were
border-zone cases, and 3 were dull-normal. One family
furnished a moron and a border-liner; another furnished a
moron, a border-liner, and a dull-normal. A moron girl in
this school has an insane mother. The girl is normally at-
tractive in appearance and has reached the stage of ado-
Room P, City " Y" This is a fourth-grade class enrolling
39 pupils, 23 of whom are over-age for their grade. Five of
these are from three to five years retarded. The ages of the
1 Compare Tables 23 and 24 with Tables 28 to 35, pp. 159-162.
124 INTELLIGENCE OF SCHOOL CHILDREN
89 pupils range from 9 years to 16 years. Of 5 suspects
tested in this room, & were feeble-minded 'and 3 border-zone
cases. Another, the lowest of all according to the teach-
er's estimate, was absent and could not be tested.
Although these three schools represent an extreme
situation, there are undoubtedly thousands of teach-
ers in the United States whose problem is made fully
as difficult by the presence of backward and feeble-
minded children. Sometimes the teacher's position is
jeopardized because of her inability to give such pupils
the expected mastery of school work. Often she is
penalized if her percentage of failure is much higher
than the average. Everywhere the emphasis is on
keeping children up to grade, rather than on finding
work which is suited to their abilities.
How many children are feeble-minded ? In " X "
County the proportion of feeble-minded children is
not far from four per cent of the total enrollment.
Fortunately this is an exceptional condition. The
proportion usually found is between one and three per
cent. In a partial survey of mental deficiency in the
schools of San Luis Obispo, California, we found two
per cent of the school children mentally defective.
The Stanford tests of 1000 unselected children in five
cities gave one per cent below 70 I Q, and two and
one half per cent below 75 I Q. Probably one and
one half per cent of the 1000 cases were feeble-minded.
Among Dickson's first-grade children the proportion
of mental deficiency was very considerably higher
MENTAL TESTS OF SCHOOL LAGGARDS 125
than this. Of Hubbard's 79 fifth-grade pupils, 4
tested below 70 I Q.
Other investigators in large number have found
similar ratios of mental deficiency. After an excep-
tionally thorough study of feeble-mindedness in the
public schools of Oakland, California, Mrs. Hicks
classifies three per cent of the children of that city
as feeble-minded. Dr. Macfie Campbell's survey of a
certain district in Baltimore resulted in a classifica-
tion of three per cent as having " pronounced mental
defect." Dr. Goddard, after a number of investiga-
tions in eastern cities, including New York City, esti-
mates that about two per cent of the school children
in any average city will be found feeble-minded.
Strikingly similar results have been found for sev-
eral rural districts. Dr. Wilhelmine Key, in a study
of a county in Northeastern Pennsylvania, finds 8.2
per cent of the population mentally defective. In a
survey of mental deficiency in Porter County, Indi-
ana, by the United States Public Health Service,
2185 children were given a Binet test. Approxi-
mately one per cent were classified as feeble-minded,
and another large group as doubtful. A similar in-
vestigation was made by the United States Public
Health Service in New Castle County, Delaware.
Abbreviated mental tests were given to all the 8798
children enrolled, and on the basis of these tests the
seriously retarded cases were sifted out for a com-
plete Binet test. As a result, 1.8 per cent were class-
126 INTELLIGENCE OF SCHOOL CHILDREN
ified as being of institutional grade, not counting
about a fifth of 1 per cent who were epileptic. We can
conclude then that on an average two or three children
out of a hundred are so poorly endowed in intellectual
ability as to render their social competency a matter of
This figure should not be surprising, considering
the number of children who are over-age three years
or more. The following per cents on this point are
3 yean 4 years
or more or more
(per cent) (per cent)
318 cities (Strayer) 5.25 1.5
Salt Lake City (Survey Report) 4.9 1.2
96 Nebraska cities and towns (Morton) 4.7 1.4
227 Michigan cities and towns (Berry) 8. ?
13,626 California rural school children (Phelps) 2.5 1.
"X" County, California (Tennan) 5.2 2.
Probably eighty per cent of those who are retarded
four years or more, and fifty per cent of those
retarded three years or more, are feeble-minded.
Many others are feeble-minded who have not at-
tended school long enough to become seriously re-
tarded. In " X " County fifty-eight per cent of the
feeble-minded were not more than two years over-age.
Criteria of mental deficiency. Certain statements
made in the preceding discussion may appear to be
based on the assumption that all children may be class-
ified as definitely normal in intelligence or definitely
feeble-minded. No such assumption, however, has
MENTAL TESTS OF SCHOOL LAGGARDS 127
been intended. The distribution of mental ability is
"continuous," by which is meant that there is no defi-
nite line of demarcation between the imbecile, the
moron, the dull, and the normal. Each group shades
into the other by imperceptible degrees. The num-
ber of individuals to be classified as feeble-minded
will depend largely on the standard of classification
used. When 75 1 Q is taken as the dividing line, the
number of feeble-minded is about two and a half times
as great as when 70 1 Q is taken. If 65 1 Q is used,
the ratio of feeble-mindedness is greatly reduced.
The different standards employed have given rise
to serious disagreements among psychologists as to
the proportion of feeble-mindedness in various social
groups. The disagreement comes from the fact that
the term " feeble-mindedness " is currently used in
two very different senses. In one sense it refers to
the possession of no more than a certain degree of men-
tal (chiefly intellectual) capacity, as measured by some
objective scale. This is the psychological definition.
As commonly employed, the term " feeble-minded "
has reference primarily to those who, because of in-
herent or early acquired mental weakness, " cannot
compete on equal terms with their fellows," or " can-
not manage themselves or their affairs with ordinary
prudence." This is the social criterion. These two
criteria, the psychological and the social, cannot be
used interchangeably for the reason that ability to get
on in the world depends upon many things besides
128 INTELLIGENCE OF SCHOOL CHILDREN
absolute mental capacity, such as health, looks, bear-
ing, muscular strength, inherited wealth, sympathetic
friends, economic and industrial conditions, the pre-
vailing level of intelligence in those with whom the
subject must compete, etc. However, experience
shows that, on any reasonable standard as to what
constitutes social competency, the outlook for chil-
dren who test below 70 1 Q is anything but favorable.
Feeble-mindedness and dullness not curable.
The classification of school children as " feeble-minded "
or " dull " can only be valid in case it is found that the
individual who tests low at an early age will continue
to test low in succeeding years. As is shown in chap-
ter IX, re-tests of children after long intervals indi-
cate that a child's brightness or dullness remains
surprisingly constant. 1 The following re-tests are
F. C. t middle-grade imbecile, tested as follows:
Age 8-6, mental age 4-0, 1 Q 47;
Age 10-8, mental age 5-4, 1 Q 50.
V. J. 9 high grade moron, tested as follows:
Age 8-6, mental age 6-0, 1 Q 71, grade 1;
Age 9-4, mental age 6-9, 1 Q 72, grade 2;
Age 11-6, mental age 8-4, 1 Q 73, grade 3;
Age 12-4* mental age 8-10, 1 Q 72, grade 3.
H. V., dull normal, tested as follows:
Age 11-0, mental age 8-10, 1 Q 80.5, grade 4;
Age 14-11, mental age 11-8, 1 Q 78, grade low 7.
1 See Figures 21 'and 22 in chapter EC, showing growth curves of
dull and feeble-minded children.
MENTAL TESTS OP SCHOOL LAGGABDS 129
Grade expectancy of the feeble-minded. Because
of the tendency of the I Q to remain constant, it is
possible to forecast with a reasonable degree of ac-
curacy the highest grade in which a dull or feeble-
minded child will ever be able to do satisfactory work.
It has been found that after the chronological age of
fifteen or sixteen years the mental age increases little
if at all. Making allowance for minor changes of a
few points in I Q we are able on this basis to make
such predictions as the following:
The child who tests at 60 I Q will in all probability
never go beyond the mental age of nine or ten years
(sixty per cent of 16 years = 9.6 years). Such a child
will never be able to do good work above the third or
fourth grade, although by the age of sixteen he is
likely to be found in the fifth or sixth grade, promoted
there because of age and size.
The child who tests at 70 1 Q may ultimately reach
a mental age of about eleven years, which corresponds
roughly to median fifth-grade ability. Such a child by
the age of sixteen may be able to do fair work in the
sixth grade, after much repetition, but is likely to be
carried by the lockstep of the school a couple of grades
beyond this. However, we have found no I Q of 70
in the high school.
An I Q of 80 means an ultimate mental age of ap-
proximately twelve and one half years. A child of the
80 class will at best be able, by the time mental
growth has ceased, to do fair or average work in the
130 INTELLIGENCE OF SCHOOL CHILDREN
seventh grade. A mechanical system of promotion
and sympathy for his retarded condition may be ex-
pected to land him in the eighth grade, or if he re-
mains in school long enough, even in the first or sec-
ond year of the high school. However, such a child
will never be able to do the work of the average high
school with any degree of satisfaction.
The child who tests at 90 is near enough the average
to make normal or almost normal progress through the
eight grades, although there is some likelihood of his
incurring retardation of a half year to a year. Such a
child, if persistent, may also be expected to graduate
from' high school, although the difficulty of making
normal progress there is somewhat greater than in the
grades below the high school, due to the fact that his .
competitors in the high school are selected pupils.
Those testing between 70 and 80 (about five per
cent of all children) compose the group which offers
the most difficult educational problem. The majority
of this group are not sufficiently subnormal to warrant
their commitment to an institution, nor are they able
to profit normally from the regular work of the school.
They furnish the bulk of those who by the age of
twelve or fifteen are two to four grades retarded. As
noticeably over-age pupils, they are the object of.
every one's sympathy. Because of the universal desire
to keep the retardation figures low, they are over-
promoted to such an extent that they are rarely able
to master their lessons. Tables 29 and 30 (pp. 159, 160)
MENTAL TESTS OF SCHOOL LAGGARDS 131
show the grade location of children testing between 70
and 79. Practically the only pupils in these tables do-
ing satisfactory work were those who were in a grade
corresponding closely to the mental age. Those whose
grade location corresponded to chronological age were
almost never doing work of average quality for the
Limitations of the special class. The remedy which
has been most often urged for the ills of the over-age
child is the special class. Although one or more such
classes are to be found in nearly all the larger cities,
the number is never sufficient to take care of more
than a small fraction -of the children who should at-
tend them. To provide special teachers enough for all
the seriously over-age children on the usual basis of
twelve or fifteen pupils per teacher is quite out of the
question. The most that the best cities have done is
only a beginning. Even if the special class were as
effective educationally as its most enthusiastic cham-
pions claim, it would still be an impossible solution of
the problem because of the prohibitive cost. More-
over, the question inevitably arises whether the ulti-
mate returns to society would not be greater if any
funds available beyond those necessary for the sup-
port of the regular classes were used to provide special
opportunity for children who are gifted.
One way to reduce the cost of special class instruc-
tion, which at present is about three times as high
as in the regular class, is to establish central schools
132 INTELLIGENCE OF SCHOOL CHILDREN
exclusively for backward children. When the pupils
are graded according to ability and type of defect, a
class of twenty-five presents a no more difficult prob-
lem than a class of fifteen which enrolls children who
are feeble-minded, epileptic, incorrigible, or physically
handicapped as well as those who are merely back-
Vocational training for backward children. How-
ever, the administrative aspects of the problem are
secondary to the pedagogical. The important task for
the school is to provide the kind of instruction suited
to the capacity of inferior minds. Whether this is done
by grouping the regular class into sections according
to ability, or by providing special classes, graded or
ungraded, does not greatly matter. The danger inhe-
rent in the present costly mode of attack is that we
may exhaust all our good will on a handful of feeble-
minded, and leave practically untouched the infinitely
larger and more important problem of providing the
dull with a kind of training which will make them so-
cial and industrial assets. The feeble-minded, in the
sense of social incompetents, are by definition a bur-
den rather than an asset, not only economically but
still more because of their tendencies to become delin-
quent or criminal. To provide them with costly in-
struction for a few years, and then turn them loose
upon society as soon as they are ripe for reproduction
and crime, can hardly be accepted as an ultimate
solution of the problem. The only effective way to
MENTAL TESTS OF SCHOOL LAGGAEDS 133
deal with the hopelessly feeble-minded is by perma-
nent custodial care. The obligations of the public
school rest rather with the larger and more hopeful
group of children who are merely inferior.
It should be clearly understood that individuals
of inferior intelligence are not necessarily undesirable
members of society. Indeed, the world has abundant
use for them. A large proportion of the tasks in the
modern organization of industries can be as well per-
formed by individuals of the 70 or 75 I Q class as by
those of superior intelligence, and with more satisfac-
tion in the performance. Mentality of eleven years is
ample for ordinary kinds of unskilled labor, and many
of the semi-skilled trades are within the reach of those
who test a year or two higher.
To make the most of this grade of ability, however,
it must be trained. (Fox children who test below 75
or 80 I Q, genuine vocational training should largely
replace the usual curriculum of the upper grammar
grades. 1 ? Nothing beyond a certain amount of relief to
the regular teacher is gained by segregating them in
special classes, unless their course of study is at the
same time vocationalized. Merely the introduction of
a little basketry or other " handwork " does not serve
the purpose. Although there are occasional happy ex-
ceptions to the rule, the average special class gives the
backward child little that will be of direct service to
him in the world. Often, indeed, it gives him little or
nothing beyond the scope of the regular curriculum.
184 INTELLIGENCE OP SCHOOL CHILDREN
The following case is a typical illustration of the
school's problem in dealing with over-age children:
M. is a Portuguese boy of 16 years. We first tested him
when he was 10 years of age. His I Q was 74. He was in
the third grade, where his work was very unsatisfactory. We
tested him again when he was 14 J years old and in the sixth
grade. At this time his mental age was 10-5 and his I Q 72.
As would be expected, his work in the sixth grade was very
inferior. By mental age he belonged in the high-fourth or
low-fifth grade. Recently M. left school at the age of 16
years, after promotion to the seventh grade. It is certain
that had M. remained in school indefinitely, he would never
have been able to master the work required for graduation
from the eighth grade. The school which he attended (a rural
school) had done all it could for him by the usual methods.
His teachers were unusually capable and conscientious. He
had been given a fair trial at the regular curriculum and, in
spite of his best efforts, for M. is an industrious lad, he
could not make headway with it. He goes out into the world
with no further equipment from his schooling than the abil-
ity to read, write, and do the fundamental operations in
arithmetic. Some children who test as low as M. would be
rated as feeble-minded. No psychologist would so classify M.
Intellectually inferior he certainly is, but as far as his intelli-
gence goes, it is sound. About ordinary affairs his judgment
is dependable, and he is steady, industrious and anxious to
make good. There are probably many lands of semi-skilled
work in which he could succeed. For none of these has he re-
ceived any preparation. After nine years in school, he faces
the world with no vocational asset but his God-given brawn.
There are approximately a million children like M. in the
public schools of the United States. 1
1 For other descriptions of dull and feeble-jninded children see.'
Tennan, Lewis M., The Measurement cf Intelligence (Houghton
Mifflin Co., 1916), pp. 82-94.
THE I Q AS A BASIS FOR PREDICTION
Prediction the essence of science. The essential
characteristic of scientific knowledge is that it can be
used as a means of predicting what will happen when
certain conditions are given. Primitive man lived
largely in a world of apparently chance events. The
progress of enlightenment is measured by man's ability
to find law and order in what seems to be but a chaos
of happenings. The sciences of physics and chemistry,
for example, have made possible thousands of predic-
tions as to what will inevitably occur in the interaction
of forces and elements, given such and such conditions.
The passage of a current of electricity through water
according to a certain procedure always gives hydro-
gen and oxygen. A bridge constructed of given mate-
rials according to given specifications will be able to
withstand a certain definite amount of strain which
can be foretold with reasonable accuracy.
The sciences that have to do with living matter, the
organic sciences, have developed more tardily than the
inorganic. The introduction of the scientific element
into economics, sociology, education, and psychology
is still more recent and far less complete. ^Education,
en -an,~ ein
basis. Teachers have been too content to believe
136 INTELLIGENCE OF SCHOOL CHILDEEN
in "miracles," instead of searching for the order
and inevitable sequence which will make prediction
possible. Some teachers even seem to prefer empiri-
cism to science, and to derive satisfaction from the
fact that child nature contains so many unknown
quantities. To such persons the proposal to develop
a science of mental growth which would enable us to
forecast a child's future may even seem repugnant. It
runs counter to the deep-seated and blind faith that
anything is possible for any child; that the material
with which education works is uniform; and that
processes alone count.
Compared to the obvious variety of the world of
adults, with its healthy and its sickly, its geniuses and
its incompetents, its moral leaders and its criminals,
the world of cradle or schoolroom does indeed present
a homogeneous aspect. But the uniformity is one of
appearance only. Mental tests are showing that the
variety is there, and that it bears certain definite rela-
tions to the variety found among adults. To the ex-
tent that differences among children are measureable,
and to the extent that these differences tend to per-
sist, prediction is possible. It is evident, therefore,
that one of the most fundamental problems of psy-
chology is that of investigating the laws of mental
growth. When these laws are known, the door of the
future will in a measure be opened; determination of
the child's present status will enable us to forecast
what manner of adult he will become. The entire
THE I Q AND PREDICTION 137
program of educational treatment which should be
accorded a given child hinges on such possibility of
Limitations of prediction in psychology. It must be
admitted, however, that the laws governing the devel-
opmentof many mental traits are still little understood,
and that tools for their exact measurement are far from
satisfactory. We shall concern ourselves here only
with the prediction of future intellectual status. The
Standardization of the Binet scale on the basis of age-
norms makes it a suitable instrument for the investiga-
tion of mental-growth curves. By applying it re-
peatedly to the same children, we can find out whether
constancy or irregularity rules. Prediction hinges on
the question whether a child who is found by the test
to be a given per cent above or below the mental level
normal for his age continues to be accelerated or re-
tarded to the same degree. The answer is found in the
extent to which the I Q remains constant.
Before presenting our data on re-tests there is one
point that should be made clear; namely, that minor
discrepancies in the results of successive tests do not
necessarily imply corresponding irregularity of mental
growth. Mental measurements are not and never will
be made with the exactness which is possible in the
physical sciences. " Accidental " and imponderable
factors are always present to invalidate the result in
some degree. This would be true even if the measuring
scale itself were perfect, for the child himself is a more
138 INTELLIGENCE OF SCHOOL CHILDREN
or less variable factor. His performance in the tests
may be influenced by conditions of health, the previous
night's sleep, fatigue, timidity, anxiety, grief, attitude
toward the examiner, or other special conditions.
When the different tests of the same individual are
made by different examiners, we have the additional
disturbing influence of the personal equation in giving
and scoring the tests.
Some have argued that such accidental influences
largely invalidate the results of mental testing. Such
persons take it for granted that an average child may
test like a dullard if he is fatigued or sleepy, and that
similar factors beyond our control may reduce the
performance of a genius to the level of mediocrity. The
question is one of fact. The results of re-tests show
that, while theoretically all these influences may be
present in some degree, their combined influence is in
most cases small.
Constancy of I Q as shown by re-tests. Re-tests
have been given to 315 children in the vicinity of Stan-
ford University. To 46 of these children three or more
tests have been given. In case of a child tested several
times, each test has been compared with each of the
others; for example, the first test with the second,
third and fourth separately, the second test with the
third and fourth separately, and the third test with
the fourth. This gives in all 485 I Q comparisons.
The re-tests were not made as a separate investiga-
tion, but are such as have accumulated as a result of
THE I Q AND PREDICTION 139
various mental test studies carried on by many differ-
ent Stanford University students over a period of
years. For this reason the tests were made under an
extraordinary variety of conditions. Thirty-three ex-
aminers contributed to the total number of tests. 1
Only twenty-eight per cent of the earlier and later
tests were given by the same individual. There was
no uniformity as regards time of day, place of testing,
freedom from distractions, etc.
The intervals between tests ranged from one day to
seven years and classify as follows:
Less than 1 year 86
1 to S years 138
3 to 5 years 85
More than 5 years 127
The ages of the subjects, counted at the time of the
earlier of two compared tests, were as follows :
3 to 5-11 99
6 to 8-11 139
9to 11-11 134
12 to 14-11 55
Above 15 8
i Irene Cuneo 148
Laura Herron 46
J.H. Williams 40
L. S. Stockton 37
Dorothy Albrecht 38
Mary B. Chamberlain 34
Lowery Howard 18
W.M. Proctor 17
Blanche Cuminings. . 11
21 examiners* 1 to 8 each 78
140 INTELLIGENCE OF SCHOOL CHILDREN
The effect of the varying ages, also of wide intervals
between tests, would presumably be to reduce correla-
tions. The normal child who is reSxamined after an
interval of five or six years earns his second mental age
on other tests of the Binet scale than those taken in
the first examination.
H the tests in the one or the other part of the scale
are not properly standardized, there will be a dis-
crepancy between the two I Q*s due to this extraneous
factor. We have therefore treated the various ages
and intervals separately. Since the tendenqy for I Q
to increase or decrease might be expected to vary ac-
cording to the brightness of the child, the I Q groups
of 89 or below, 90 to 109, and 110 or above, have
also been treated separately.
Table 25 shows the frequency of various amounts of
I Q change in the different groups of children classified
(1) according to interval between the tests, (2) accord-
ing to age at the time the earlier test was given, and
(8) according to brightness. Increases in I Q of later
as compared with earlier tests are tabulated as +
changes, decreases as changes.
Table 25 shows that it makes little difference
whether the child was bright, average, or dull, how
long an interval separated the tests, or what the age
of the child was at the earlier test. The majority of the
changes are for all groups relatively small. The salient
facts for the entire series of re-tests may be summa-
rized as follows:
THE I Q AND PBEDICTION
Age at first test
I Q Group
TABLE 25. SHOWING I Q CHANGES FOR CHELDEEN
AJTBR DOTTEBENT INTEBVALS, FOB CHJLDBEN OP
*** *"" -"*" ^ TT ^^ T\--rvni xv
142 INTELLIGENCE OF SCHOOL CHILDREN
(l)-Tke central tendency of change is represented by an in-
crease of 1.7 in I. Q ;
() The middle fifty per cent of changes lies between Hie
limits of 3.3 decrease and 5.7 increase ;
(8) The probable error of a prediction based on the first test
is 4.5 points in terms of I Q.
A more impressive way of expressing the agreement
between earlier and later tests is by means of a correla-
tion array, as is done in Table 26 for all the tests taken
together. The correlation is .933. Those who ranked
high in the earlier test ordinarily ranked high in the
later, the average remained close to average, the low
Personal equation of the examiner. If an intelli-
gence scale yielded consistent results only when used
by the same examiner its value would be extremely
limited. On the other hand, if results secured by dif-
ferent examiners in testing the same subjects give a
high correlation, a most important requirement of
validity has been met. Separate tabulation of those
cases in which the earlier and later tests were made
by different examiners yielded a correlation of .929,
almost exactly the same as that for all the cases com-
pared without regard to examiner. The following are
typical illustrations :
Re-tests were made by Mir. and Mrs. Stockton of
forty children who had been previously tested at vari-
ous ages by various examiners. When the records were
compared with the original tests, it was found that in
. .r-l i-l-^O lO^tCQ t-l
144 INTELLIGENCE OP SCHOOL CHILDREN
twenty-five of the forty cases the I Q had not changed
more than four points, and in thirty not more than six
points. The correlation with the earlier tests was .94.
Twelve children of Fresno, California, who were ex-
amined by Dr. J. H. Williams in 1915 were reSxamined
by Miss Blanche Cummings. Dr. Williams was trained
at Stanford University, while Miss Cummings had
learned the Stanford Revision procedure by studying
the directions in The Measurement of Intelligence. The
results of the twelve repeated tests are shown in
The coefficient of correlation between the tests of
Dr. Williams and those of Miss Cummings, made over
three years apart, is .96, Spearman method. Not only
do the tests agree with each other; the school progress
of the child agrees with both. The average or su-
perior children make normal or more than normal
progress; the inferior children less than normal pro-
gress. Other groups re-tested by different examiners
have given similar results.
Influence of interval between tests. Table 25
shows that it makes little difference whether the com-
pared tests are separated by an interval of a few
months or several years. The central tendency of
change and the proportion of changes included in a
given range remain much the same. The only excep-
tion is that tests separated by more than five years
show a greater tendency toward increase of I Q than
is the case with shorter intervals. This is probably a
THE I Q AND PREDICTION
11 High School
I High School
I High School
n High School
TABLE 27. SHOWING AGREEMENT BETWEEN THE EARLIER AND
LATER TESTS OF TWELVE CHILDREN
Second test after an interval of three years. (Correlation .964)
spurious result due to the fact that in case of inter-
vals of this length the first test was made by a form
146 INTEmGENCE OF SCHOOL CHILDREN
of the Binet scale differing slightly from that used in
the later tests.
It is rather surprising that children profit little in a
re-test from their experience in the first test. One
would naturally expect a considerable improvement
due to their feeling more at ease and to the opportu-
nity to think over their earlier mistakes and correct
them. However, this advantage yields the child (on
the average) only two or three points in I Q even
when the test is repeated within a few days. 1
Influence of brightness and dullness on the con-
stancy of the I Q. There is a widespread popular
opinion that bright children usually fail to hold
their own, and that the dull are likely to improve
with increase of age. Psychologists have more often
expressed the view that it is the dull who fail to
hold their own, and that the superiority of the bright
probably increases. Table &5, which gives the I Q
changes separately for the bright, average and dull
(above 110, 90-109, and below 90), shows that the
I Q remains almost equally constant for the three
groups. The central tendency of change for the bright
is + 0.7; for the average, + 3.0; and for the dull,
+ 1.2. The greatest tendency to gain appears with
the average group, and the next greatest with the dull.
The differences, however, are practically negligible.
1 For further data on re-tests of kindergarten children see the arti-
cle by Irene Cuneo and Lewis H. Terman, Pedagogical Seminary,
THE I Q AND PREDICTION 147
When those above 125 I Q and those below 80 I Q
were treated separately, the central tendency of
change was found to be 0.5 for the former, and
+ 1.2 for the latter. The very dull actually gained a
trifle more than the very bright. However, we have
only thirty-one repeated tests for the low group to
compare with eighty for the high group, and the low
group contains very few cases below 60 I Q. It is
possible that feeble-minded children testing below 60
are less likely to hold their own than those of milder
degree of defect. As far as the school is concerned
this possibly may be ignored, since there are relatively
few in public school classes who test this low.
On the other hand, the I Q as determined by the
Stanf ord-Binet (or any other intelligence scale yet de-
vised) cannot indefinitely maintain its constancy in
the case of children who are exceptionally superior.
The child of fourteen years who tests at 139 has passed
all the tests in the scale. Thereafter his I Q drops
gradually to 122, which is the maximum possible for a
subject of sixteen years who passes all the Stanford-
Binet tests. Similarly, the child who tests at 161 has
reached his maximum I Q at the chronological age of
twelve years. This does not mean that his develop-
ment ceases at this time, but merely that the Stanford'
Binet does not measure it. Children who test at 180
are measured fairly accurately up to the age of fifteen
years, or nearly as far as chronological age is counted.
Since only about one child in a hundred rates as high
148 INTELLIGENCE OF SCHOOL CHILDREN
as 130, the scale is seen to offer a reasonably satisfac-
tory measure for ninety-nine per cent of unselected
children, and also for the remaining one per cent
except during the later years of mental growth.
Limits of accuracy in prediction of mental develop*
ment. From the frequency of the various amounts of
change in I Q, as shown in Table 25, we can compute
the average error which will be made in predicting the
mental age or the I Q which a child will have at any
later age. Speaking roughly, fifty per cent of the I
Q*s found at a later test may be expected to fall within
the range between six points up and four points down.
Half of this distance, or five points, is the probable
error of an I Q for purposes of prediction. 1 Deviations
of one, two, three, four, or five times the probable er-
ror may be expected to occur with the frequency given
in the second column below. 2 The frequency actually
found is shown in the third column.
Deviations as great as Theoretical Actual
or greater than frequency frequency
(per cent) (per cent)
Onetime P. E. ( 5 points) 50 50
Two times P. E (10 points) 27.76 16.6
Three times P. E. (15 points) 4.3 6.2
Four times P. E. (20 points) 7 1.85
Since the central tendency of change is toward an
1 Actually the P. E. is somewhat less than this, namely 4.5.
2 See any textbook on statistical method; e.g., Rugg, H. O., Sto-
tistical Methods Applied to Education (Houghton Mifflin Co., 1917);
THE I Q AND PREDICTION 149
increase of a little more than 1 point, and since the
changes above and below this are distributed fairly
symmetrically, we may say, roughly speaking, that
the chances that an I Q will either increase as much as
six points or decrease as much as four points are one
in two; that it will either increase as much as twelve
points or decrease as much as eight points, one in five;
that it will either increase as much as eighteen points
or decrease as much as twelve points, one in twenty;
that it will either increase as much as twenty-four
points or decrease as much as sixteen points, one in a
hundred and forty.
The above statements regarding the probability of
different degrees of change occurring include devia-
tions both above and below the central tendency of
change. The chance for a deviation to occur in one di-
rection isonly half as great. For illustration, the chance
that a child who tests at 85 will later test as high as 91
is one in four; that he will later test as high as 97, one
in ten; that he will later test as high as 103, one in
forty, etc. Similarly, the chance that he will drop to 81
or below is one in four; that he will drop to 77 or be-
low, one in ten; that he will drop to 73 or below, one in
forty, etc. 1
It is evident, therefore, that the I Q is sufficiently
constant to make it a practical and serviceable basis
1 The chances of given deviations occurring vary so slightly for
children of different ages at the time of the first test, also so slightly
for intervals of different length between tests, that for practical pur-
poses these factors may be left out of account.
150 INTELLIGENCE OP SCHOOL CHILDREN
for mental classification. At the same time it is not
infallible. A single test does not give us certainty, but
merely a high degree of probability. While the I Q
it yields is extremely valuable in the tentative classifi-
cation of children, it needs to be checked up by sup-
plementary data and by re-tests. In certain types
of pathological subjects the I Q may undergo large
fluctuations. Epileptics, for example, frequently de-
teriorate from something like normality to middle-
grade deficiency in the course of a few years.'
Mechanical errors as a source of discrepancy. So
dose is the agreement in most cases between earlier
and later tests that when a discrepancy of more than
twelve or fifteen points is found it warrants a strong
suspicion that an incorrect age has been given in one
of the tests, or that arithmetical error has been made
in adding credits to find mental age or in dividing
mental age by chronological age to find the I Q.
Mistakes of this kind are a more dangerous source of
error than the personal equation of the examiner.
Arithmetical errors can be greatly reduced by making
all computations twice, a precaution which we con-
sider absolutely necessary. The avoidance of errors
due to incorrect age is by no means easy. Chil-
dren in the lower grades occasionally do not know
their age. Sometimes the age recorded in the school
register is incorrect because of falsification by parent.
The seriousness of this source of error is shown by the
THE I Q AND PREDICTION 151
K. N. tested at the mental age of 5-2. The age given was
6-6, and the I Q was therefore computed as 79. This indi-
cated a degree of dullness almost amounting to borderlinity.
However, when the child was re-tested two years later the
chronological age was given as only 7-6, instead of 8-C.
Investigation disclosed the fact that 7-6 was correct and
that the parents had falsified the age to secure earlier en*
trance. The mental age earned at the second test was 7-2
and the I Q 95. Correction of the age at first test raised the
former I Q of 79 to 94, practically the same as that earned
in the second test.
Do adenoids affect the I Q ? It is very generally
believed that adenoids seriously retard mental devel-
opment, and that their removal is nearly always fol-
lowed by a marked intellectual awakening. If such
were the case the effect of removal should be to in-
crease considerably the I Q. Among our re-tested
children we have records of twenty-seven who under-
went an operation for removal of adenoids or tonsils
in the interval between tests. Comparison of the I Q's
of earlier and later tests showed a central tendency
toward a gain of two points and a fraction. There
were ten losses and seventeen gains, but no gain larger
than fourteen points, and only two larger than ten
points. Although these results are too scanty to war-
rant a conclusion, they suggest that adenoids and
diseased tonsils may give a child an exaggerated ap-
pearance of dullness. They are a chronic source of
toxins which .seriously impair physical vitality, and
their removal probably adds to the child's vivacity
152 INTELLIGENCE OF SCHOOL CHILDREN
and to his interest in school work. This effect would
easily be mistaken for real intellectual improvement.
There are enough reasons why adenoids and diseased
tonsils should be removed, apart from any effect on
Investigations on a larger scale should be made to
determine the effects on intelligence not only of ade-
noids, but also of such factors as malnutrition, chorea,
loss of sleep, fatigue, hookworm, 1 malaria, etc.
9 10 u a
FlG. 20. MBNTAlXJBOWTHCUBVBSASTH]reWOUIJ)BBIPlQwinBB
1. Mental-growth curve as it would be for a child who continued to test at 100 1 Q
2. Mental-growth curve as it would be for a child who continued to test at 183 1 Q
S. Mental-growth curve as it would be for a child who continued to teat at 671 Q
1 An investigation made by Strong of the effects of hookworm
disease on mental and physical growth did not afford very positive
results, as re-tests were not given.
THE I Q AND PREDICTION
Curves of mental growth. If we had a perfect scale
for determining the mental age level, and if the I Q re-
mained absolutely constant, the " curves " of mental
growth would be a straight line from birth to the point
of mental maturity. The mental-growth curves for
typically dull, average, or bright children would then
be as represented in Figure 20. It will be observed
that each of the hypothetical growth curves in Fig-
ure 20 maintains a certain relative distance from the
heavy line representing the average-normal (I Q 100).
FIG. 21. ACTUAL MENTAL-GROWTH CURVES OF CHILDREN OF
VARIOUS DEGREES 03 BRIGHTNESS
1. Mental-growth curve of superior child No. 17, Chapter XI
8. Mental-growth curve of superior child No. 11, Chapter XL
4. Granddaughter of an inventor and related to John Wesley
5. Son of a man of international fame
13. Member of a feeble-minded family of "X" County, California, which has long
burdened t^ community with delinquents fypd paupers
154 INTELLIGENCE OP SCHOOL CHTLDBEN
The child's brightness or dullness is not at all indi-
cated by his mental age, but only by the ratio of mental
age to chronological age. The tendency is to remain
a certain per cent above or below the normal.
We do not have an infallible measuring scale, and
even if we had we should hardly expect the I Q (i.e.,
the ratio of mental age to chronological age) to main-
tain perfect constancy. Accordingly, mental-growth
curves can only be expected to agree roughly with
those shown in Figure 20. Figure 21 shows actual
mental-growth curves found by repeated tests of chil-
dren of various degrees of brightness.
t i 9 4 9
r ft 9 x>
FIG. 22. MENTAL-GROWTH CURVES OP BRIGHT AND DULL CHILDREN
2. Superior child No. S3, Chapter XI
4. Described in The Measurement qf Inlelligaue, page 97
6. Enrolled in a special class. In the third grade at 12} years
7. Enrolledinaspecialclass. In the fifth grade at age of 16, but unable to do the work
THE I Q AND PREDICTION
Figure 22 contrasts two groups of children. Those
below the normal line are all either feeble-minded or
border-line cases, most of whom were attending spe-
cial classes and none of whom was ever able to pro-
gress above the seventh grade. The lowest is an imbe-
cile, barely able at the age of twelve years to read in
the first reader. The bright group is as much above
average intelligence as the dull group is below, but
they attract far less notice in school.
Figure 23 illustrates how children in the same fam-
ily ordinarily test close together. The mental-growth
curves above the normal line represent two brothers
ft ft 14
FlG. 23. MBNTAl>GROWTHCimyBSINTwoCONTRASTmGFAMIIJia
1. , and S. Two brothers and one water of the W. fainfly
4, 5, 6, and 7. Three brothers and one sister of the P. family
The cross represents a single test
The children of the two families attend the same school
THE I Q AND PREDICTION
brother and sister, the broken lines, 3 and 4, two
brothers in another family. The two crosses indicate
single tests of two brothers in a third family. In each
case the contrast in school success was as marked as
the contrast in the growth curves.
Figure 25 shows four exceptionally irregular curves
of mental development. No. 1 and No. 4 represent
conditions of mental disease (dementia precox and epi-
lepsy) . Normal children do not often show as marked
irregularity as that found in No. 2 and No. 8.
The I Q as a basis for predicting school progress.
The relative permanency of the I Q enables us to pre-
FIG. 25. FOTJB EXCEPTIONALLY IBBEGULAB GBOWTH CURVES
1. A boy who at the age of ten showed symptoms of dementia jmscox
. A normal boy cause of the irregularity not known
8. An adolescent boy of marked instability and neurotic symptoms
4. An epileptic girl, showing typicel epileptic deterioration
158 INTELLIGENCE OF SCHOOL CHILDREN
diet with some degree of approximation the mental
level a child will attain by a given age. We have also
seen that it is the mental level, more than anything
else, which determines a child's proper location in the
school grades. If schools were careful to grade children
according to mental age it would be possible, knowing
a child's I Q, to predict in what grade the child would
be found at any given time in the future.
We have seen, however, that schools do not grade
children as nearly by mental age as they should. While
children of low I Q do become retarded, they are nev-
ertheless usually found in a grade considerably above
that corresponding to mental age. On the other hand,
children of very superior I Q, while they are likely to
be promoted somewhat beyond average children of
their age, are usually found in a grade considerably
below that corresponding to mental age.
Notwithstanding this constant tendency of teachers
to promote children by age rather than by ability, the
I Q nevertheless offers a fairly serviceable basis for
predicting a child's later school progress. Tables 28 to
35 show the grade which children of various degrees of
brightness had attained at various ages. The heavy
squares running diagonally across each figure show
the grades in which a child of 100 I Q normally be-
longs at the various ages. In all these tables, age
seven includes children between six and one half and
seven and one half years old, age eight those between
seven and one half and eight and one half, etc.
THE I Q AND PREDICTION
ra iv v vi
VII Vm Total
TABLE 28. GRADE PROGRESS AT 50-69 I Q
ra iv v vi
VII VIII Total
TABLE 29. GRADE PROGRESS AT 70-74 I Q
160 INTELLIGENCE OF SCHOOL CHILDREN
I II HI IV V VI VII Vm Total
TABLE 30. GRADE PBOGEESS AT 75-79 1 Q
m iv v vi vn
TABLE 31. GRADE PBOGEESS AT 80-84 I Q.
THE I Q AND PBEDICTTON 161
VI VII Vm Total
TABLE 32. GRADE PBOGRBSS AT 95-104 I Q
HI IV V VI
VII VIII Total
TABLE 33. GBADE PROGRESS AT 120-129 I Q
Inspection of the above tables will reveal the follow-
1. The lower the I Q, the greater the degree of re-
162 INTELLIGENCE OF SCHOOL CHILDREN
i n nr iv v vi vn vin Total
TABLE 34. GRADE PROGRESS AT 130-139 1 Q
I H III IV V VI VII VIII Total
TABLE 35. GRADE PROCESS AT 140-170 1 Q
tardation. As we go from the 50-69 1 Q group to the*
95-104 group, the grade location gradually improves
until it approximates the normal.
2. The I Q groups above 100 show a greater degree
of acceleration the brighter the group. It will be noted,
THE I Q AND PREDICTION 168
however, that the acceleration of the bright group is
not quite as great as the retardation of the dull. For
example, the 120-129 children are not as far above
the heavy squares as the 70-74 and 75-79 children are
below. The same holds for the 180-139 group as
compared with those of 60-69 1 Q.
3. If the mental age of a given child in one of the
retarded groups is computed, it will usually be found
that the child is less retarded than he ought to be.
When the mental age of a child in one of the bright
groups is computed, it will be found that the child is
less accelerated than he ought to be.
4. The typical child of 60 or 65 I Q tends to remain
in the first grade until the age of ten or eleven years,
and not to reach the fifth grade until the age of four-
teen or fifteen years. By this time he has a mental level
of only about nine years and is not able to do the
school work satisfactorily above the third or fourth
5. The typical child of 75-79 I Q reaches the fifth
grade by the age of thirteen years, and if he remains in
school is likely to be found in the eighth grade by the
age of sixteen or seventeen. Nearly always, however,
his grade location is higher than the mental age would
6. Children of 80-84 I Q usually remain two years
in the first grade, and complete the eighth grade, if they
complete it at all, one or two years behind schedule
164 INTELLIGENCE OF SCHOOL CHILDBEN
7. On the other hand, children of 12O-129 I Q are
usually found either one or two grades accelerated.
Nearly all of this gain, however, is made in the first
year or two of school life. After the first year, they
are held to the one-grade-one-year pace of average
children. Even so, the central tendency is for them
to complete the eighth grade at the age of thirteen.
8. The situation is slightly but not proportionately
better for the I Q group of 130-139. Children of 140 to
170 I Q, however, are likely to become three or four
years accelerated and to reach the eighth grade by the
age of eleven or twelve years. Wherever children of
the higher I Q groups are located, their work always
presents a striking contrast with that of children of
the 60, 70, or 80 I Q class who are several years their
SOME FACTS ABOUT FIFTY-NINE SUPERIOR
Educational neglect of superior children. The at-
tention of teachers is constantly being called to the
large number of defectives among school children, and
to the educational and social problems to which they
give rise. For the intellectually superior, however,
the ones upon whose preservation and right education
the future of civilization most depends, no special pro-
vision is made. In the average school system then*
very existence, even, is ignored. Yet, as we have
seen, they are just as numerous as the dull and
mentally defective. The latter attract attention by
their inability to do the work and by their malad-
justment to school discipline. Children of superior
ability are often submerged with the masses simply
because they are not recognized.
Another thing that has blocked the educational
path of the gifted child is the widespread belief that
intellectual precocity is pathological, that exception-
ally bright children are usually unhealthy and likely
to become physical or mental wrecks if their intellec-
tual interests are at all stimulated. Recently, how-
ever, the truth of the traditional belief has come more
1 Written with the assistance of Margaret Hopwood Hubbard.
166 INTELLIGENCE OF SCHOOL CHILDREN
and more under suspicion. Such studies as have been
made of gifted children have not disclosed the patho-
logical symptoms popularly supposed to accompany
exceptional intellectual ability. On the contrary,
wherever the experiment has been tried of provid-
ing such children larger and richer opportunities the
results have been surprisingly gratifying. , When fa-
vored with extra promotions they make good in the
higher grade without injury to health; when given
the advantage of a broader and richer course of study
their minds expand and take the wide swath as easily
as they had taken the narrow one.
Instances of this kind coming to our attention
from time to time led us some years ago to undertake
the more or less systematic study of exceptionally su-
perior children. With the help of Margaret Hop wood
Hubbard and other Stanford University students, we
have secured Binet tests of some eighty California
children having an I Q above 135. All but a few
of these tested 140 or above. Fifty-nine of the group
were subjected to a rather careful study, which will be
summarized briefly in this chapter.
Selection of subjects. The study was limited
chiefly to children who tested 140 or above in order to
secure subjects whose intelligence would be as far above
average as that of typical feeble-minded children is
below average. The fifty-nine subjects composed
two groups, which will be designated as the Alameda
Group and the Miscellaneous Group.
FACTS ABOUT SUPEBIOR CHILDREN 167
The Alameda Group included twenty-four subjects
selected by systematic search throughout the public
schools of Alameda, California. The method of se-
lection was as follows: (1) The age-grade location was
found for all the children in the grades below the high
school; (2) the pupils were rated by their teachers for
intelligence on a scale of five: very superior, superior,
average, inferior, and very inferior; (3) all the children,
who were under-age two or more years by the Ayres
standard, and who were rated by their teachers as
above average in intelligence, were provisionally se-
lected for study; (4) the principals and teachers were
asked to recommend others of exceptionally superior
intelligence who could not qualify on the above rules. 1
In this way seventy children of the five thousand
enrolled in the schools of Alameda were provisionally
selected and given a test with the Stanford-Binet
1 The fifty-seven Alameda pupils who were tested but fell below
the standard of brightness set for the study, included twenty-four
boys and thirty-three girls, with I Q's as follows:
130-135 9 8
125-129 4 8
120-124 2 4
115-119 1 6
110-114 3 4
105-109 1 3
Total 24 33
The two who tested below 100 were each one year over-age for
their grade. The teacher's judgment was in error because age had
not been taken into account.
168 INTELLIGENCE OF SCHOOL CHDLDBEN
scale. Of these, twenty-two were found to have an in-
telligence quotient of one hundred and thirty-seven or
above. In addition, one subject of 135 1 Q and one of
136 I Q were included for special reasons, making a
total of twenty-four in the Alameda Group.
The Miscellaneous Group consisted of twenty-five
children who had been located by the writer and by
various Stanford students in the half-dozen years
preceding the present study.
Supplementary data. The intelligence tests were
used primarily to identify the superior children, and
to measure their degree of superiority. Much more
time was given to interviews with parents and teach-
ers in the work of gathering items of information
listed in an eight-page " information blank " for par-
ents, and a somewhat shorter one for teachers. All
of the homes were visited by Mrs. Hubbard except a
few of the more distant ones. The interviews lasted
usually one to three hours. Similar interviews were
held with the child's teacher.
For the Alameda Group, duplicate office records
were secured from the school principals, showing each
child's school marks from the time of entering school.;
A number of educational measurements were also
available for comparative purposes in the case of this
group, including tests in addition, subtraction, multi-
plication, division, spelling, and arithmetical reasoning*
The information blank for teachers called for (1)
data on school progress; () ratings of the quality of
PACTS ABOUT SUPERIOR CHILDREN 169
the child's school work in each subject (usual scale of
five) ; (3) indications of exceptional ability, and a state-
ment as to whether the ability appeared to be special
or general; (4) ratings on twenty mental, moral, and
physical traits (scale of five used for each trait) ; (5)
facts regarding play, reading, physical defects, nerv-
ousness, eccentricities, moral peculiarities, etc.
The information blank for parents called for data on
(1) nationality, education, and occupation of parents;
names and ages of all the children, with rating of each
for intelligence; (2) ratings of the superior child on the
same twenty traits which were rated by the teacher;
(3) facts regarding walking, talking, dentition, nour-
ishment in infancy, early growth, illness, etc.; (4) spe-
cial data on adenoids, tonsils, eye and ear defects,
headaches, digestive trouble, nervousness, timidity,
amount and quality of sleep; (5) regularity of school
attendance, attitude toward school, home study and
reading, use of time after school, evenings and vaca-
tions, private instruction, etc.; (6) indications of supe-
rior ability, amount and kind of formal instruction in
the home, vocational ambitions; (7) occupation, edu-
cation, and ability of parents and grandparents, and
data regarding uncles, aunts, cousins, and distin-
guished or defective relatives.
The traits which were rated both by parents and
teachers were: studiousness, power to give sustained
attention, persistence, social adaptability, leadership,
initiative, evenness of temper, emotional self-control,
170 INTELLIGENCE OF SCHOOL CHILDREN
physical self-control, will power, cheerfulness, cour-
age, sense of humor, obedience, conscientiousness, de-
pendability, intellectual modesty (lack of vanity),
unselfishness, general health, and general intelligence.
Intelligence quotients. The I Q's of the fifty-nine
subjects were distributed as follows:
The average I Q was 149*7 and the median 145.
Only eighteen were as high as 150. l
1 Since this study was completed, twenty-one other children have
been located in California who test above 140 1 Q, bringing the total
number to eighty. As many of these as possible will be followed up
from year to year until adult life. The twenty-one new cases not in-
cluded in the present study are as follows:
Sex Age Grade IQ
7-10 5 174
FACTS ABOUT SUPERIOR CHILDREN 171
The lowest of our subjects is probably equaled or ex-
ceeded in brightness by not more than 1 child in 100;
the highest by not more than 1 in 10,000 or 0,000.
The highest I Q found in Alameda's enrollment of
5000 was 158. An I Q of 140 probably occurs with an
average frequency of about 1 in 200, or one half of one
per cent. It was found nineteen times in Alameda's
enrollment of 5000, giving a ratio of a little less than
one half of one per cent.
The average I Q for the children of the different
ages was as follows:
Age Number of Average
^ Subjects I(?
3 1 162
4 2 143
5 2 144.5
6 2 144
7 4 158
8 4 147
9 10 147.4
10 10 154.5
11 11 14S.4
12 6 141.8
13 4 140 '
14 3 139.3
There were few subjects above the age of twelve
years, because the search was confined almost entirely
to the grades below the high school. The diminishing
number below eight is explained by the difficulty
teachers find in recognizing the superior child until he
has attended school two or three years.
Of the fifty-nine, only eighteen were girls. Of the
172 INTELLIGENCE OF SCHOOL CHILDREN
eighteen testing 150 or above, only two were girls. The
six highest were boys. During the progress of the work
effort was made to ignore the possibility of sex differ-
ences. Of the seventy Alameda children selected for
testing, more than half were girls. 1 Of the six girls test-
ing above 145 I Q, all but one (a child of three years)
have special ability in literary or artistic lines; the
fields in which women have met the most pronounced
Age-grade location. Counting a child at-grade who
is in the first grade between the ages of 6-6 and 7-5, in
the second grade between the ages of 7-6 and 8-5, etc.,
we have the following distribution:
On baas of Number Percent
At grade 4 8.5
Advanced one year 14 29.8
Advanced two years 14 29.8
Advanced three years 9 19.2
Advanced four years 6 12.8
Judged by appearances, the above showing is re-
markably good, for 61.8 per cent of the children are
advanced two years or more. On the basis of menial
age, however, the showing is strikingly different
Taking as our standard for the first grade the mental
age of 6-6 to 7-5; for the second grade, mental age
7-6 to 8-5, etc., we have the following:
1 Among the twenty-one superiors discovered after this studj
was made, there was a larger proportion of girls. One of these testec
FACTS ABOUT SUPERIOR CHILDREN 173
On bads of
Retarded five years
Retarded four years
Retarded three ye^r s ....
Retarded two years ,
Retarded one year ,
. . . 8
Reckoning on the basis of actual age, we find an
average acceleration of slightly more than two years;
on the basis of mental age, an average retardation of
about .6 years. The story is plainly told in Tables
36 and 37.
i n m rv v VT vn vni ix x" xi xn
6-7 to 7-6
7-7 to 8-6
8-7 to 9-6
9-7 to 10-6
10-7 to 11-6
11-7 to 12-6
12-7 to 13-6
13-7 to 14-6
TABLE 86. SHOWING HOW SUPERIOR CHILDREN ABE ABOVE-GRAB*
ON THE BASIS OF CHRONOLOGICAL AGE
Teachers' ratings on qualify of school work. The
children were graded by their teachers on a scale of
1, 2, 3, 4, 5 (very superior, superior, average, inferior,
and very inferior), in each of the school subjects.
Each child's ratings in the several subjects were then
174 INTELLIGENCE OF SCHOOL CHILDREN
i ii in iv v vi vii vin ix x xi xn
6-7 to 7-6
7-7 to 8-6
8-7 to 9-6
10-7 to 11-6
11-7 to 12-6
12-7 to 13-6
13-7 to 14-6
14-7 to 15-6
15-7 to 16^
16-7 to 17-6
17-7 to 18-6
18-7 to 19-6
TABLE 97. SHOWING HOW SUPEBIOB CHILDREN ABB BELOW-
GBADB ON THE BASIS OF MENTAL AGE
averaged. The lowest average rating for any child
was 2.91, or slightly better than "average" for all chil-
dren. The highest was 1.16. The 461 ratings in the
individual subjects were distributed as follows:
Number Per cent
1, very superior '227 49.3
2, superior 133 28.8
3, average 80 17.4
4,inferior 15 3.2
5, very inferior 6 1.3
The six ratings as low as 5 (** very inferior "), were
distributed one each in music, spelling, nifl,n\ifl,1 train-
FACTS ABOUT SUPEBIOR CHILDBEN 175
ing, and language, and two in writing. The fifteen rat-
ings of 4 (" inferior "), were distributed as follows:
music three, spelling one, manual training two, drawing
four, nature study one, writing four. There are no
grades below 8 (" average ") 9 in arithmetic, reading,
history, geography, or deportment.
Although these children averaged about two years
above grade, the ratings show that they were doing
work of a decidedly superior quality. No wonder,
since they were still located in grades below their
mental levels. There is every reason to believe that
they would continue to do superior work if they were
promoted to the grades where they belong by mental
age. Three had been so promoted, and their average
school marks were 1.44, .08, and .55. Some of these
.children and also many other superiors whom we have
tested have received promotion as a result of our
recommendations, and we have yet to find a child
who failed to make good.
Educational measurements. In connection with
another investigation 1 all of the public school children
of Alameda above the fourth grade were given the
Courtis Arithmetic, Ayres Spelling, and Stone Rea-
soning tests. Eighteen of the twenty-four of Alameda
were tested at the same time.
The tests revealed the following interesting facts:
1. The average scores of these superior children
were higher than the average of any of the grades of
* By Mr. 0. S. Hubbard.
176 INTELLIGENCE OP SCHOOL CHILDREN
the city with the exception of the high-eighth. In
addition and spelling they even excel the high-eighth.
. Of the six pupils in the low-fifth grade, four were
above the eighth-grade median in addition and sub-
traction, two above the eighth-grade median in multi-
plication, two in spelling, and one in division.
3. One girl, aged 10-2, low-fifth grade, I Q 148,
practically doubled the score of the high-eighth grade
in addition, subtraction, and multiplication, and did
considerably better than the eighth grade in division
4. Two of the six pupils in the low-fifth grade made
scores in arithmetical reasoning about 50 per cent
higher than the city's median for the high-seventh
5. In arithmetical reasoning, the subject which
more than any other taxes the real mental ability of
the pupils, the average score of the eighteen pupils was
nearly two grades above the city average for the
grades in which they were located.
Entering age and rate of advancement Of the
forty-nine subjects who had entered school and for
whom data were available, seven entered at five years,
twenty-four at six years, seventeen at seven years, and
one at eight.
Of the seven children who started to school before
six years, two skipped half of the first grade, two
others the third, one the third and seventh, one the
fourth, and one skipped two grades not designated.
PACTS ABOUT SUPERIOR CHILDREN 177
Of the twenty-four who entered at six, four skipped
the first grade, and seven others skipped half of the
first grade. Of the seventeen entering at seven, only
seven had attended the first grade, three having en-
tered at once the second grade, five the third grade,
one the fourth grade, and one the sixth grade.
It is often argued by teachers that children who are
allowed to skip grades will later be handicapped by
gaps in their knowledge. Our data show how little
truth there is in this view. Nearly all of these children
had skipped one or more grades, yet their school work
was in most cases so superior as to suggest the desira-
bility of additional promotions. Gaps in training are
quickly filled. Of course, it would be better still if
school children were so classified as to permit superiors
to make maximum progress by continuous -rapid speed
without the necessity of skipping.
According to the statements of the parents, of the
fifty who were in school, thirty had been allowed by
their parents " to go their own pace." Thirteen had
been mildly encouraged by their parents to make rapid
progress and to excel in their school work. Seven
had been purposely held back by the parents, a few
because of ill health, others in the belief that preco-
cious mental development is something to be pre-
vented as far as possible. In only two cases had there
been any serious attempt in the way of intensive mind
culture at an early age. 1 " Only" children would ordi-
1 See children No. 34 and No. 39, chapter xi.
178 INTELLIGENCE OF SCHOOL CHILDREN
narily be expected to get more than their share of early
instruction, but only seven of our fifty-nine cases
were only children, and only two of these had an I Q
Age of learning to read. Learning to read consider-
ably in advance of the normal age of six is a significant
indication of superior ability. It is ordinarily not un-
til the mental age of six years that children are able
to learn to read as first-grade children are normally
expected to do. The child of four years who learns
to read as readily as the average child of six, will
almost certainly test as high as 150. Several of the
children who did not learn to read before six wanted
to learn earlier, but were discouraged from doing so.
The one who learned to read latest, between seven
and eight, was said to have shown a desire to learn to
read at four years.
Records were obtained for forty-nine, as follows:
Between 2 and 3 years, 1, or 2 per cent
3 and 4
4 and 5
5 and 6
6 and 7
7 and 8
Attitude toward school work. Both parents and
teachers were questioned regarding the attitude
toward school work. Of the fifty for whom data
were secured, forty-three were said to like school very
much, three fairly well, and three not particularly well.
FACTS ABOUT SUPERIOR CHILDREN 179
One of these was a boy who was taught at home
and did not enter school until the age of eight years,
just after the death of his mother. He entered at
once the sixth grade, the grade where he belonged by
mental age; but the physical restraint in the class
suited to twelve-year-olds, who had been in school six
years, was naturally irksome to a young child who had
always been acccustomed to unlimited freedom. The
teachers did not understand his case and for a time he
was very unhappy. He has since been taken from the
public school, and placed in a private school where
the discipline is less exacting. As a result, his attitude
has undergone a radical change. There were equally
good explanations in the other two cases.
All but five were very regular in school attendance,
and these had missed time only because of illness.
These five were rather delicate, yet in spite of illness
and frequent absence they stood at the head of their
Play and recreation. It is generally believed that
children of exceptional intellectual ability are likely to
have little interest in play. We sought information
on this point from the teachers, rather than from
the parents, in order to secure an impartial judgment
based on a knowledge of many children. Of fifty-
one for whom data were secured, thirty-eight were
described by both teachers and parents as entirely nor-
mal in their play. Seven of the.others were said to play
less than average children, but to play normally when
180 INTELLIGENCE OF SCHOOL CHILDREN
they do play. Three had always been alone, and so
preferred to play alone. One was too timid and re-
served to mingle well with other children, and two
others were said not to care to play with children of
their own size. The abnormalities of play life do not
appear to be more numerous or more serious than
would be found in any group of children picked at ran-
dom from the school population.
The data on out-of -school activities showed that our
superiors were accustomed to spend their time after
school like average children, playing, practicing music
lessons, doing chores, running errands, etc. Satur-
days were usually spent in the same way, with per-
haps a dancing lesson, a hike, or a gymnasium period
in addition. Many had gardens which they cared
for, several had paper routes, others regular work in
a store or elsewhere.
Twenty-eight were taking private instruction in
music, twelve dancing lessons, and four language
lessons. Only twenty were not receiving private
instruction of some kind. The time devoted to pri-
vate lessons, including practice, ranged from two
to fourteen hours per week, the average being 5.3
The time devoted to home reading ranged from two
to twentynone hours per week, with an average of 7.6
hours. The books given as samples of the children's
reading were classified as "good** or "mediocre."
Needless to say, most fell in the former group. Among
FACTS ABOUT SUPERIOR CHILDREN 181
the books and authors most frequently named were
Stevenson, standard books of history, Dickens, Mark
Twain, Cooper, geographical books, nature books,
Conan Doyle, biographies, Eugene Field, Shakes-
peare, books of travel, Irving, Scott, Ben Hur, Jack
London, Little Lord Fauntleroy, Black Beauty 9 Arabian
Nights, Alice in Wonderland, Pilgrim's Progress, Bo6-
inson Crusoe, the Odyssey, the Iliad, Greek myths,
Book of Knowledge, ASsop's Fables, Bible Stories,
books on science and mechanics, and such magazines
as Youth 9 s Companion, American Boy, Harper 9 s, St.
Nicholas, and Literary Digest. Several had evidenced
a strong liking for encyclopaedias and dictionaries.
Trait ratings. Parent ratings on the twenty traits
were secured for fifty children, and ratings by both
parent and teacher for forty. It was taken for granted
that the parent-ratings would be too high. On the con-
trary, the average for the parent-rating is lower than
the average for the teacher-rating in the case of nine-
teen of the twenty traits. This is probably explained
by the fact that the parent is compelled to take as a
standard of comparison the child's brothers, sisters,
cousins, or friends. To the extent that abilities are
hereditary this would tend to give a higher standard
than that employed by teachers, whose classes are
composed of children of all grades of ability.
Table 38 shows the individual traits arranged in
two rank orders; first according to teacher-ratings,
then according to parent-ratings.
182 INTELLIGENCE OF SCHOOL CHILDREN
Geneml intelligfiflcfc t T , , T . .
Dependability .......... t -
Studioiisness ...... T , . * -, -
Cheerfulness. ........ ^ . t ,-
ConscfentioTispess ..-.- T -
Unselfishness t. T .- f -.-T. T -
Wi,<w? of hunior ...........
Evenness of ten^pei* ,,.. -,
Intellecfaiftl modesty, ......
Emotional self -control
Initiative . . - -, T .-
TABLE 38. SHOWING AGREEMENT BETWEEN THE TRAIT EATINGS BY
TEACHEES AND PARENTS. (Correlation between the two rankings
The above ratings show that the parents and teach-
ers agreed closely on the traits in which these children's
superiority is most marked. General intelligence, sus-
tained attention, will power, persistence, dependabil-
ity, and studiousness, ranked high in the estimate of
both parents and teachers; social adaptability and
leadership lowest. Parents and teachers differed great-
est in their ratings on the following:
Unselfishness parents rate higher
Courage teachers " "
!Emotional self-control teachers ** **
Obedience teachers ** **
FACTS ABOUT SUPEBIOR CHILDREN 183
In studiousness, cheerfulness, and general intelli-
gence, no child was graded below & by either teacher or
parent. In power to give sustained attention, persist-
ence, will power, conscientiousness, and dependability,
only one child was marked lower than 3. In power to
give sustained attention, persistence, and will power,
the mark 4 was given to three children. In courage and
sense of humor only three of the fifty children were
marked 4, inferior. In initiative only two children were
marked 4, and two 5. In unselfishness four children
were marked 4. Five of the children were below aver-
age in evenness of temper, five in intellectual modesty,
five in general health, seven in physical self-control^
eight in emotional self-control, ten in social adaptabil-
ity, and eleven in leadership. This is not far from the
number we would expect to find below average in an
ordinary group of children.
There are two reasons why a rather large number
are graded below average in social adaptability and
leadership: (1) in school most of them are associated
with children who are older and whose greater physi-
cal maturity gives them an advantage over our young
and inexperienced subjects in play activities. Six of
the children rated 4 in social adaptability by the
teacher were rated either 8 or 2 by the parent, which
would indicate that among their own special friends
and playmates they were average in these traits.
(2) In a few instances the child of superior mental
ability does not care to play. His preference for read-
184 INTELLIGENCE OF SCHOOL CfflLDBEN
ing and private lessons prevents him from becoming
a leader or an important member in a social group.
However, IS of the 89 ratings in leadership were
1 (14.6 per cent), and 18 of the ratings in social adap-
tability (20.2 per cent). Five were rated 1 in social
adaptability by both teacher and parent. The pro-
portion of leaders is probably larger than would be
found in a group of unselected children.
Moral traits. About half the twenty traits on which
our subjects were rated might be classified as moral
traits. " Obedience," " conscientiousness," " dependa-
bility," "unselfishness," " evenness of temper," and
"will power "belong very definitely to this group. The
average rating for the children as a group on these
traits was as follows:
Obedience 2.16 1.51
Conscientiousness 1.94 1.61
1 Dependability 1.92 1.56
Unselfishness 1.42 1.73
Evenness of temper 2.22 1 .90
Willpower 1.81 1.50
In all of the moral traits except " unselfishness," the
teachers 9 ratings were higher than those of parents.
The children were rated higher by their teachers in
deportment than in a majority of their studies. The
average rating on deportment was 1.54, a record
equaled by only three of the school studies, and not
considerably exceeded by any.
These ratings would indicate that our subjects are
FACTS ABOUT SUPERIOR CHILDREN 185
about as superior morally as they are intellectually.
Additional information pointing in the same direc-
tion was obtained in response to the following request
in the teacher's blank: "Describe any moral faults or
peculiarities such as disobedience, obstinacy, dishonesty,
selfishness, inability to get on with others, unusual or a&-
normal sex interests, lack of balance, etc." Data were se-
;ured for fifty-three children. Of these, forty-six were
said to have no moral faults or peculiarities worthy of
mention. Of the remaining seven, one " takes pleasure
in others' mistakes," one has " a rather bad disposi-
tion," one cries very easily, one is obstinate and " lacks
will power to make himself do the things he does n't
like " (certain of the school work), one girl is " very
much interested in boys," and another girl is " shy"
and " reticent." Practically all of these faults are such
as would hardly be thought deserving of mention in
the case of average children. They stand out in these
children by contrast with their general superiority in
other traits. There is only one child in the entire group
who appeared to be seriously lacking along moral lines. 1
The typical superior is exceptionally lovable and
charming, the kind of child one would like to adopt.
Health and physical traits. The average rating by
parents on general health was 2.14; by teachers, 2.10.
There were only two ratings of 5 ("very inferior")
and two of 4 (" inferior "). On the other hand, there
were twenty-eight ratings of 1 ("very superior").
1 See child No. S3, chapter XL
186 INTELLIGENCE OF SCHOOL CHTLDBEN
Only four were said to have defective vision, and
only one defective hearing. Twenty-one had under-
gone operation for removal of adenoids, and two others
were known to have more or less adenoid trouble. The
record for tonsils was similar. The fact that approxi-
mately half of our superior children have had either
adenoids or diseased tonsils suggests that these defects
may not be as injurious to mental development as
common opinion would have us believe.
One had chorea a few years ago, but has recovered.
Two others had noticeable muscular twitchings.
There were two stutterers in the group, both of whom
at the time of the investigation were taking corrective
lessons. There were no cases of abnormal fears. A
part of the nervousness and restlessness occasionally
mentioned was probably due to their not having
enough school work to keep them busy. One boy,
asked how he liked school, said he liked it in the
morning but not in the afternoon, because by noon
he always knew his lessons and then there was nothing
to do! So much has been said about the nervous un-
balance of precocious children that it is surprising to
find over two thirds described as free from symptoms
of this kind. The symptoms of most of the others in-
dicated nothing serious. The proportion of stutter-
ing and chorea was not far from that which is usually
found for unselected children.
All but three of the children were said to sleep " per-
fectly ." The average time of sleep for the children of
FACTS ABOUT SUPERIOR CHILDREN 187
each age was found to be slightly greater than the
Terman and Hocking averages for 69 unselected
school children. 1 There was no case of marked sleep
Of the nine who were said to have occasional head-
aches, eight have them very seldom, not more than
two or three times a year. One had long been subject
to serious recurrent headaches.
Five were described as " not strong." One of these
had always been sickly and at the age of eight years
had only attended school one year. In that year, how-
ever, he did the work of the first three grades. Another
of these has also had insecure health from birth. He
did not enter school until the age of fourteen. Be-
tween the ages of six and twelve he had only one hour
per day of private instruction, and in that time com-
pleted the work of the first eight grades. The other
three of the five were apparently just not strong
enough to endure serious physical strain or excitement.
Only three were seriously handicapped by ill health, a
record which would probably not be excelled by an
equal number of school children picked at random.
Table 39 gives the ranges for age of walking and talk-
ing in comparison with those for Mead's two groups
of normal and feeble-minded children.
1 Terman, Lewis M., and Hocking, Adeline: "The Sleep of-
School Children; Its Distribution According to Age and Its Relation
to Physical and Mental Efficiency " ; Journal of Educational Psychol-
ogy (1913), pp. 138-47; 199-208; 269-82. See also chapter xx of L.
M. Terman's Hygiene of the School Child for a digest of this study.
188 INTELLIGENCE OF SCHOOL CHILDREN
Mead's normal group*
Mead's feeble-minded group
Superior children group
Mead's normal group
Mead's feeble-minded group
Superior children group ..........
TABLE 39. AGE OF WALKING AND TALKING FOR SUPERIOR CHILDREN
AS COMPARED WITH FEEBLE-MlNDED AND NORMAL
* C. D. Mead: The Relations of General Intelligence to Certain Mental and Phy-
sical Traits. Teachers College, 1916; pp. 117.
The average age of learning to walk is a little more
than half a month lower for our superiors than for
Mead's normals, and nearly eleven months below the
average for his feeble-minded. The difference in aver-
age age of learning to talk is greater, our superiors
being three months ahead of Mead's normals and
twenty-three months ahead of his feeble-minded.
Social status and heredity. We have classified our
children according to the occupational status of the
fathers, basing the classification upon Taussig's five
occupational groups. Our subjects classify as follows:
Class 1 81, or 53 per cent
Class 2 22, or 37 " "
Class 3 6, or 10 " "
The results indicate that parents of a grade of in-
telligence low enough to keep them in the unskilled
FACTS ABOUT SUPERIOR CHILDREN 189
or semi-skilled class are not likely to produce children
of the grade of ability represented in this study. Of
the seventeen subjects testing above 150 1 Q, sixty-five
per cent belonged to class 1, thirty-five per cent to
class 8, and none to class 8. ; Several children of the
two lower social groups were brought to our attention
and were tested, but in no case was the I Q above 130.
There is a tendency on the part of teachers to over-
estimate the intelligence of such children. The labor-
er's child of 130 1 Q attracts about as much notice as
a college professor's child testing at 150.
Information was sought regarding the child's
brothers and sisters, parents, grandparents, cousins,
uncles, aunts, and any other relatives of superior ability.
Twenty-nine of the parents mentioned relatives whom
they considered superior. Fifty-one superior uncles,
thirty-seven superior aunts, and numerous cousins and
remote relatives were mentioned. The large major-
ity of the children had at least one grandparent known
to be a superior. Among the more remote ancestors
mentioned were Whistler, Edwin M. Stanton, Samuel
Adams, Roger Williams, Colonel Crawford, Ralph
Waldo Emerson, Stonewall Jackson, John Hancock,
Hancock Jackson (a governor of Missouri), and Arch-
bishop Tait. Others whose names were not given
were designated as " a sculptor," " an artist," " a me-
chanical genius," " an eminent man in the South dur-
ing the Civil War," " a president of a western college,"
" an inventor," and " an exceptional musician."
190 INTELLIGENCE OF SCHOOL CHILDREN
That the parents of our superior children were
themselves superiors is further indicated by the extent
of their education. Of the one hundred and twelve
parents for whom data were available, fifty-two
(46.4 per cent) were college graduates, and ninety-one
(81.2 per cent) were graduates of a secondary school.
In the population at large, the proportion of college
graduates is probably not more than one fortieth as
high, and the proportion of high-school graduates
probably not more than one tenth as high, as that
found for the parents of our superiors. Of the one
hundred and twelve parents, sixteen (14.3 per cent)
had done post-graduate work in a college or university.
Of the one hundred and seventy-two grandparents
for whom data were secured, seventy-two (42.4 per
cent) were graduates of a secondary school, while
twenty-three (18.4 per cent) were graduates of a col-
lege or professional school. When we consider the
limited opportunities for higher education at the time
when these grandparents were youths, this record is
hardly less remarkable than that of the parents. It
is evident that most of these children had sprung from
a decidedly superior stock.
Does the superiority tend to disappear? Excep-
tional brightness in children is often regarded as merely
a matter of precocious development, the assumption
being that the final level attained is ordinarily no
higher than in the case of children who test at average-
normal. This assumption finds no support in any of
FACTS ABOUT SUPERIOR CHILDREN 191
the exact observations that have been made. Several
studies have shown statistically that children who
make exceptionally good records in the lower grades
also as a rule make superior records in the high school,
and the correlation between high-school grades and
college grades has also been found to be positive and
We have had a number of superior children under
observation for six to eight years, and in no case has
there been any indication of a tendency toward deteri-
oration to the level of average. If there were any con-
stant tendency toward deterioration this should reveal
itself in a decrease of the I Q with increase of age.
However, re-tests of superiors show that the I Q is
more likely to increase. Of our fifty-nine superiors,
nineteen have been tested two or more times. The
greatest loss in the re-tests was 10 points, while the
greatest gain was 21 points. The central tendency
was toward a gain of .08 points. 1 (See, for example,
Figures 21 and 22, pp. 153 and 154.)
The results of the re-tests are corroborated by an-
other line of evidence. One year after this study was
made, the parents of the fifty-nine superiors were
asked to re-rate the children on each of the twenty
traits, and to give detailed information regarding any
changes that had occurred in health, social adaptabil-
1 However, as superiors of 180 1 Q or above become adult the I Q
rating fails to do them justice, as the highest 1 Q possible for an adult
is 122 by the Stanford-Binet. Children of 140 1 Q are not adequately
measured above thirteen or fourteen years. (See p. 147.)
192 INTELLIGENCE OF SCHOOL CHILDREN
ity, quality of school marks, and ease with which
school work was done. Replies were received for
fifty-one children. For no child did the average rating
on the twenty traits show any considerable change
from that of the year before. The gains and losses
were all slight, and almost exactly balanced each other.
The results on health, school marks, ease of carrying
school work, and social adaptability were as follows:
Bet- The Not so
ter same good
Health 8 44 4
School marks 5 40 6
Ease of school work 2 47 2
Social adaptability 9 42
On the whole, the amount of change appears well-
nigh insignificant. Such changes as occurred in social
adaptability, which constitutes the greatest problem
for superiors, were all in the direction of improvement.
There is one circumstance which tends to make the
superiority of bright children less apparent (but not
less real) with increase of age. In the lower and mid-
dle grades all the children attend school, and the supe-
rior child in these grades is compared with the average
for children in general. In the upper grades the chil-
dren of inferior ability are rapidly eliminated, and here
the superior is compared with " survivors " who com-
pose a highly selected group. For this reason, the
child who is correctly rated as " very superior " in the
fifth grade may rank as merely "superior" in high
school, and perhaps as only " average M in college. He
FACTS ABOUT SUPERIOR CHILDREN 193
has not deteriorated; the average for his class has
Conclusions. The data which have been presented
in this chapter justify the following tentative conclu-
1. That intellectually superior children are appar-
ently not below the average in general health;
2. That in the vast majority of cases their ability is
general rather than special or one-sided;
8. That the superiority is especially marked in
moral and personal traits;
4. That " queerness," play deficiency, and marked
lack of social adaptability are the exception rather
than the rule;
5. That while superior children are likely to be
accelerated on the basis of chronological age, they are
usually two or three grades retarded on the basis of
6. That their school work is such as to warrant
promotion in most cases to a grade closely correspond-
ing to the mental age;
7. That the superiority tends to show early in life,
is little influenced by formal instruction, and is per-
8. That superior children usually come from superior
CASE STUDIES OP FORTY-ONE SUPERIOR CHILDREN
THUS far our discussion of superior children has been
impersonal and statistical. We have sought to find
out what is true of such children in general, as regards
their physical, mental, and moral traits and the in-
fluences which would explain them. This chapter will
be devoted to brief descriptions of typical cases, in
order that teachers may see in a concrete way what
the superior child is like and sense the pedagogical
implications of his presence in the school. The case
studies to be presented could easily have been expanded
to the length of a chapter for each child, and such de-
tailed descriptions would be of the greatest interest.
Our present purpose, however, permits only summary
treatment of the most salient facts regarding a limited
number of typical cases.
Most of the children to be described belong to the
group discussed in the foregoing chapter. Apart from
the results of the Binet test, the data to be set forth
were in most cases furnished by teachers and parents.
Their statements have been in part summarized and
in part quoted, though usually with abbreviations
and with omissions of matters of secondary interest.
The "trait rating" mentioned is always the average
rating given (by teacher or parent) on
FORTY-ONE SUPERIOR CHILDREN 195
physical and mental traits named on p. 182. It will
be remembered that in these ratings 1 is "very su-
perior," "superior," 3 "average," 4 "inferior," and
5 "very inferior." It will be noted that in no case
does the average rating of a child on the twenty traits,
either by parent or teacher, fall as low as 3.
,i No. 1. Boy: E. M. 1 Illustrating exceptionally rapid
school progress and unusual will power.
First test: age 6-11; mental age 10-0; I Q 145; not
Second test: age 7-10|; mental age 13-8; I Q 167;
Third test: age 10-0; mental age 16-7; I Q 166; high-
In the second examination (age 7-10) E. passed the
induction test, 2 the arithmetical reasoning and the
clock test in year 14, the code test and six digits back-
ward in Average Adult, and repeated eight digits
direct order and seven digits reverse order in Superior
When E. was tested at the age of 6-11 and earned
an I Q above 140, the prediction was made that he
would be able to enter high school at the age of 11
1 See brief description of E. M. at age of eight, in Terman, Lewis
M. The Meamrement of Intelligence (Houghton Mifflin Co., 1916),
* These different tests are all described at length in Terman's 2%e
Measurement of Intelligence, which see.
196 INTELLIGENCE OF SCHOOL CHILDREN
years, or possibly at the age of 10. He did so at the
age of 10-5, having completed the work of eight grades
in three years. Practically every mark, except in
handwork, has been perfect. He entered the first
grade at 7? years. On the first day of school he was
placed in the first grade, but within an hour he was
promoted to the second grade, by noon to the third,
and to the fourth before the end of the school dajfc
His teacher had studied exceptional children and was
able to recognize superior ability. Under the average
teacher it would probably have taken E. two years
instead of one day to reach the fourth grade.
E.'s father is a professional man: the mother a uni-
versity graduate and formerly a teacher. The mater-
nal grandmother was a university graduate and a
school principal of more than ordinary ability in math-
ematics. E.'s ability in mathematics is also marked.
Parents 9 notes. Health record good. Ability is
fairly general, but somewhat special. More marked
along mathematical and scientific lines than others.
Wonderfully adept at arranging and classifying facts.
When between three and four years of age could add
long numbers. Learned to read at the age of 5 by
following his mother around and asking the names of
letters, and soon afterward surprised his parents by
reading fluently out of a primer. Has had no formal
home instruction, but parents have been careful to
answer all his questions. Does little studying at home
and reads only about seven hours per week. Spends
FORTY-ONE SUPERIOR CHILDREN 197
his spare time in play, delivering papers, etc. Excep-
tionally dependable and takes life seriously. Helps
his father a great deal in the office, and can be safely
entrusted with important responsibilities in the de-
tails of office work. Has sometimes to be kept out of
such work because of worrying about getting it through
promptly and accurately. Average parent rating on
Later report, age 10, eighth grade. Health excel-
lent. Reads more and does less statistics. Work in
manual training still mediocre. School marks excel-
lent, but not quite as good as formerly. "He seems
to have his hands full for the first time in his life."
Growing more adaptable and agreeable. Elected to
various offices in school. Parents' rating on traits at
this time, 2.08, about the same as before.
Teacher's notes. Ability not altogether even. Spell-
ing and arithmetic perfect. At the age of 8 did the
work of the eighth grade in mental arithmetic tests.
Has a wonderful memory for facts, but does not often
ask for reasons or explanations. " Ability above aver-
age in all lines, but especially so in statistics, facts, or
anything capable of formal array. Can tell the study
and recitation schedule of every class, and remembers
the lesson assignment for all the other pupils; can tell
who missed certain words yesterday in any class."
Rather enjoys the mistakes of others. Exceptionally
calm and quiet. Teacher's average rating on traits,
198 INTELLIGENCE OP SCHOOL CTTTLDBEN
One of the most interesting things about E. is the
fact that his school record has been better than that
of many other superior children testing fully as high.
To one who knows him the reason is dear. His will
power and determination are about as superior as his
intelligence. He will not allow any one to excel him
in mental work. In manual training, however, his
work is inferior even to that of the average child of his
No. 2. Boy: H. B. Illustrating extreme retardation
in school, although nearly as bright as No. 1.
Age 8-7; mental age 18-10; I Q 150; low-third grade.
Vocabulary score, 44. Passed the box test and re-
peated six digits backward in Average Adult.
With a mental age of nearly 13 years, H. is in the
grade which corresponds to his actual age of 8j. His
mother wants him advanced because, she says, " He
gets so tired of school when he finds it so easy to keep
ahead of his class." However, he has only been in
school one year and has been allowed to pass through
two grades in five months.
Parents 9 notes. Was seriously ill for some time in
his first year. Health now good, except for occasional
digestive trouble. Slightly nervous.
At the age of 5j read like an average pupil in the
second grade. At 7 read everything from children's
books to newspapers and magazines, reading every
FORTY-ONE SUPERIOR CHILDREN 199
word and understanding the text. At 5 years read
numbers to the thousands. At 5j counted to a thou-
sand. No special instruction beyond answering his
questions in a simple, truthful and thorough manner.
Has unusual ability in oral expression. Average
parent rating on traits, 2.00.
Later, age 9|, in the high-fourth grade. Health
good. Greatly interested in the progress of the war,
inventions, conversation, etc. Doing well in piano
lessons. He ranks with one other pupil as the best in
his class. No home work. At ease in any group and
evidently a natural leader. Average teacher rating
on traits at this time, 1.20, or considerably better than
Teacher's notes. " He is very musical. His mental
ability, however, is general. Says he expects to * know
lots of things/ Would read continually, if permitted."
No. 3. Boy : A. W. Illustrating the value of men-
tal tests in school grading.
Age 5-8; mental age 7-6; I Q 133; not in school.
Age 6-8; mental age 8-8; I Q 180; second grade.
At the age of 5-8 passed all but the vocabulary test
in year 8, arranged the weights in year 9, and passed
the three-word test in year 10.
As a result of the test the father, a superintendent
of schools, was urged to send the boy to school at once
and to see whether he would not be able to complete
200 INTELLIGENCE OF SCHOOL CHILDBEN
the first two grades in one year. A few months later
the father wrote as follows : " A. is learning to read very
rapidly. In four weeks he has learned to read the
entire primer of 187 pages. Four weeks ago he could
not read a line." A year later, at the age of 6-8, A.
was leading his class in the second grade, and at the
age of 7 years was doing splendid work in the low-third
grade. The father writes at this time, " A. seems more
interested now than ever. School marks excellent
and the work perfectly easy."
It is altogether probable that but for the test and
for the fact that the father was superintendent of
schools, and therefore able to secure extra promotions,
A. would have gone through school without ever hav-
ing an opportunity to do work commensurate with his
No. 4. Boy: 8. 8. Brother of R. S., No. 5. Il-
lustrating exceptional mental balance. Later devel-
Age 4-7; mental age 6-8; I Q 145.
Age 5-10; mental age 8-9; I Q 150.
Age 7-0; mental age 10-8; I Q 153.
At the time of the third test S. had not yet started
to school. Vocabulary score at the age of 7, 28 words. 1
At this time passed four tests in year 1$, including
abstract words, ball and field, fables and similarities.
1 Indicating a total vocabulary of approximately 5000 words.
FORTY-ONE SUPERIOR CHILDREN 01
In the induction test, year 14, announced the rule
governing number of holes before the end of the ex-
periment, but was unable to double 16.
The following notation appears on the record for the
first test, when S. was 4j years old: cc By 8 years S. will
test llj." His test at the age of 7 gave him a mental
age of 10-8, so it appears the prediction will be more
Parents 9 notes. No serious illness except the ordi-
nary children's diseases. Has always shown remark-
able power of reasoning. Has had little home instruc-
tion, but is reading in the second reader (age 7). Is
omnivorous as to the books he wants read to him.
French lessons twice a week and some instruction on
the piano. Allowed to go his own pace. " However,
we have always answered his questions truthfully and
fully. We have always allowed him to take the initia-
tive, have never suggested his memorizing anything,
have never forced anything on his attention.*' Early
ambition was to be a railroad engineer. Recently he
cherishes the hope of becoming a reformer! Average
parent rating on traits, 2.00.
S. is a most lovable boy, quiet and retiring yet not
bashful. His bearing is one of very modest dignity.
He is perfectly unspoiled. Father a college professor
of journalistic experience; mother a college graduate of
unusual ability and marked musical talent. Several
relatives of superior ability, S. developed much earlier
302 INTELLIGENCE OF SCHOOL CHILDBEN
in childhood than his sister (who tests at 147) and gave
somewhat more evidence of superior ability*
No. 5. Girl: R. S. Sister of S. S., No. 4. Artis-
tic ability and marked emotionality. Underrated by
Age 4-10; mental age 7-1; I Q 147; not in school.
Wonderfully responsive. Pull of life and the picture
of health. Talked most charmingly and with utter
lack of self -consciousness all the way from her home to
the laboratory where she was to be tested. Although
less than 5 years old, she passed the test of arranging
weights in year 9. The parents were greatly surprised
that her I Q equaled that of her brother. They had
probably not made sufficient allowance for the differ-
ence in age.
Parents 9 notes. R/s aptitude is described by her
parents as being in the direction of artistic expression.
"She sings wonderfully true to time and key and
dances with natural grace. She has acquired a sure-
ness of stroke in drawing which an equal amount of
Montessori training never gave her brother. She has
natural dramatic ability, but lacks the development of
abstract thinking which characterized her brother.
She has never been asked to learn anything, although
her questions have always been answered fully and
truthfully. However, she has never asked as many or
varied questions as her brother, from whom she has
FORTY-ONE SUPERIOR CHILDREN 208
learned most of what she knows." Average parent
rating on traits, 2.85. Obedience and emotional self-
control were both rated 5. R. is said to be emo-
tional, impatient, and inclined to fly into fits of
screaking if things displease her. Play life normal.
No. 6. Boy: J. 8. l Lovable disposition. Indica-
tions of literary ability.
Age 8-2; mental age 11-4; I Q 138; high-fourth
Age 11-0; mental age 15-; I Q 136; high-seventh
Age 12-3; mental age 17-7; I Q 144; high school.
J.'s I Q is by no means as high as many others we
have found, but he has such a winning personality,
charming disposition, and uniform ability that we
consider Trim one of our most promising superiors.
The father was a man of superior ability, and the
mother had been secretary of a large business firm.
Both parents died several years ago, and J. has been
reared by his aunt. On his twelfth birthday J. handed
his aunt a beautiful letter which he had written, on
his own initiative, to express his appreciation of the
way she had cared for him. This is typical of his
loving and lovable disposition.
1 See brief description of J. S. at the age of 8 years, in The
Measurement of Intelligence, Lewis M. Terman (Houghton Mifflin
Co., 1916), p. 99.
204 INTELLIGENCE OF SCHOOL CHILDREN
J.'s unusual talent for writing is shown by the fol-
lowing poems written before his eighth birthday.
They are reproduced without change of spelling or
Hurrah for Christmas
And all it's joys
That come that day
For girls and boys.
Flowers in the garden.
That is all you see
Who likes them best?
That's the honey bee. '
My mother 9 s busy
My mother is very busy today
And all I have to do is play.
If I only knew what she had to do
I'd like to help her, would n't you?
What a trouble washing day;
It seems my mother can never play.
I wonder if she'll get tired out
From walking, walking all about.
Here is Sunday, resting day;
That's the best tiring I can say.
We go to church and pray and pray,
That's the hardest thing I say.
FORTY-ONE SUPERIOR CHILDREN 205
Before the age of eight J. amused himself by writing
fables to which he always attached a correct moral. 1
The following is a sample:
A Fable The Selfish Boy and the Poor Girl
Once there was a rich boy in a city and he went into a
candy store and bought some candy. When he came out he
still had a lot of money. While he was walking down the
street he met a little girl selling shoe laces. He just kept
on eating candy and did not buy anything from her or even
offer her a piece of candy.
About a month later the rich boy's house was robbed and
this little girl was getting a lot of money. The boy now had
to go around selling and he met the girl many times, but she
never helped him because when she had been poor he did
not help her.
Moral: Those you do not help will not help you.
No. 7. Boy: T. B. All-round ability, with special
interest in medicine. Musical family.
Age 10-5; mental age 15-2; I Q 146; sixth grade.
Vocabulary score 64, which is practically median
for mental age 16. Passed the ingenuity test in
Father French, mother American. A great grand-
uncle was Meyerbeer, the French composer. Another
uncle is a locally well-known violinist and composer.
Parents 9 notes. T. has always been perfectly healthy
* It will be remembered that the Stanford-Binet test of fable
interpretation brinjp an average of only two successes for five fables
at the age of twelve, and four successes for five fables at the Average
206 INTELLIGENCE OF SCHOOL CHILDBEN
except for slight nervousness. Somewhat myopic.
Learned to talk at 7 months. School work easy; does
little home work except in the practice of music, of
which he is very fond. Shows a remarkable interest
in medical science. All his childish games and all his
reading have tended in this direction almost from the
time he could talk. " Have tried to hold hi back
because of his tender age and temperament/' Al-
though healthy, he has always been high strung. Chief
indications of superiority, his passionate desire to
learn and his obsession for medicine.
Teacher's notes. School work excellent, except
drawing. " T. expresses his thoughts on any subject
in a marvelous way for a boy of his age. He is capable
both in his oral and written work. Very studious and
interested in his work. His power of attention some-
times seems lacking, but when I have called it to
his attention on certain occasions he has said, ' I was
only day-dreaming.' " Very adaptable socially. Abso-
lutely unspoiled. Very conscientious and unassuming.
Enjoys reading medical works, especially in the sur-
gical line. Beads from a medical encyclopaedia. Also
studies electricity and likes to experiment. Very
strong sense of truth and marked straight-forwardness .
T. is probably one of the most promising of our
superiors. His interest in medicine was evident in the
sixty-word test, in which he gave the names of numer-
ous bones, muscles, and other organs of the body. We
FORTY-ONE SUPEBIOR CHILDREN 207
have here not a case of one-sided ability, but a mind
of very superior general ability focussed upon a special
No. 8. Boy: P. T. Ordinary parents and dull
Age 11-11; mental age 17-7; I Q 148; low-eighth
This boy is specially interesting because of the con-
trast with his brother, who at the age of 6-10 tests at
mental age 5-8 ; I Q 88. The parents say the two chil-
dren are absolutely unlike, and the verdict of the tests
agrees with this opinion.
The father is a carpenter. Neither parent has had
more than a common school education, but the mother
is somewhat above the average in intelligence. A dis-
tant relative of the mother was a high official in the
Methodist Episcopal Church, and a relative of the
father was an archbishop of Scotland.
Parents 9 notes. P. shows unusual ability in all of
his school work and also in music. He succeeds in
everything he undertakes. When he was 22 months
old he knew the names of the important buildings in
San Francisco and could point them out on a photo-
graph of the city. Was never taught at home beyond
the alphabet. Health record good. Desires to be-
come a mechanical engineer. The younger brother
expects to be a fanner. Average parent rating on
208 INTELLIGENCE OP SCHOOL CHILDBEN
One year later, age 18. Excellent record continued
in every respect. This time the mother rates the child
1 on every trait. She is probably realizing more and
more the contrast with the younger brother.
Teacher's notes. All-round ability. " A. is a great
reader and a most satisfactory pupil." Teacher's
rating was 1 on all but two of the traits.
No. 9 and No. 10. C. D. and L. D. Brother and
sister. Exceptional children of ordinary parents.
C. Age 14-6; mental age 19-0; I Q 181; third year
L. Age 10; mental age 1S-8; I Q 137; high-fifth
C. made the remarkable vocabulary score of 82
words, which equals that of the average university
senior. He has reached a stage of development where
the Stanford-Binet falls short of being an adequate
A brother of C. and L. is in the seventh grade at the
age of 11, and a sister is in the second grade at the age
of 7. Neither has been tested, but both are said to be
as bright as C. and L.
In one respect this is the most interesting family of
children of whom we have record. The father is a
barber, the mother was a tailoress before marriage,
and not a single known relative has had more than a
common school education or intelligence above the
ordinary. Each of the four children belongs to a grade
FORTY-ONE SUPERIOR CHILDREN 209
of superiority not encountered of tener, on an average,
than once among one hundred children.
Parents 9 notes. C/s health is perfect except for
myopia and slight headaches. Entered second grade
at the age of 6 and shortly afterwards skipped to third.'
Spends all his spare time in reading. Learned the
alphabet at years and could read books and news-
papers at 8. Special ability in mathematics. No
special instruction, but has been encouraged.
L. is more sociable, talkative, and active than C.
Her health is very good and her school work gives her
no trouble. She is less studious than C. but gave in
childhood similar indications of superiority.
Teacher's notes. The teacher says regarding C.'s
high standing in class, "I would cite as evidence of
unusual talent his answers to questions proposed dur-
ing the lesson, which are almost invariably in a sin-
gle short sentence covering completely the ground."
Social adaptability inferior. Is pensive, very shy,
and retiring in a crowd of boys. Remarkable power
L. is described by her teacher as exceptionally quick
and accurate in her work and alert to everything.
#0.11. Boy:B.F. 1 Remarkable all-round ability,
which was greatly underestimated by the parents.
i See brief desertion of B. F. at the age of 8 years, in Terman,
Lewis M., The Measurement cf Intelligence (Houghton Mifflin Co.,
1916), p. 102. At that time B. F. was the brightest child in the
210 INTELLIGENCE OF SCHOOL CHILDREN
First test: age 7-8; mental age 12-4; I Q 161; high-
third grade. Vocabulary score in this test was 40
(median for 12 years). The induction test in year
14 and the box test in Average Adult were both
Second test: age 9-4; mental age 15-7; I Q 167; low-
sixth grade. In the second test the vocabulary score
was 56. The fables, box and code tests of Average
Adult, and the paper-cutting test and abstract pas-
sages of Superior Adult were passed.
B.'s father is an able minister, and the mother is a
woman of exceptional intellect and personal qualities.
The following statement by the parents illustrate how
the superior child in a superior home is likely to be
underrated because of the high standard by which he
is judged: "His development has seemed to us quite
normal and even. We had not thought of him as
much above the average in intelligence/' (Mother.)
"Really Mrs. F. and I think that some mistake has
probably been made in the observations upon which
your rating is based. While B. is an alert, good and
thoroughly satisfactory boy, we have never thought
of him as considerably above the average in mentality.
We have tried to be good parents to him, provoking
inquiry, answering questions and giving him oppor-
tunity for a variety of experiences that would furnish
raw material for his ideas." (Father.)
Nevertheless, B. has a grade of intelligence which is
FORTY-ONE SUPERIOR CHILDREN 211
probably not equaled by more than one child in five
Parents 9 notes. B. has never been seriously ill, but
there is a slight tendency to stutter when he is excited.
Learned to read at 5. At 7 read Hiawatha and com-
mitted 150 lines of the poem to memory. Does no
home study and reads only about a half hour per day.
Spends hours after school in outdoor play, marbles,
football, and base ball; also practices on the piano.
Has an unusual fund of information in history and
current events. Catches impressions easily. Many
interests. Allowed to go his own pace in school, and
has had no formal instruction in the home. Wants
to go through college and become a minister. Takes
his place well among other children without being a
Two years later, age 11. The mother writes that
notwithstanding a change of schools the teacher con-
siders B. ready for the eighth grade. School work
easier than ever. Shows a growing interest in world
problems. Average parent rating on traits at this
Teacher's notes. Unusual ability in reasoning and
an exceptional fund of general information. Also
considerable ability in music. In two and a half years
has almost completed six grades. Remarkably attrac-
tive and alert. Not particularly handsome, rather
delicate in appearance, but vigorous in his play and
a favorite with the children who congregate in his
212 INTELLIGENCE OF SCHOOL CHILDREN
yard. Rated 1 by the teacher on every one of the
twenty traits, the only one of our superior children
with whom this occurs.
At the time of the first test B. was above 12 years in
mentality, but was in the high-third grade. On the
showing made in the test we urged the father to try
to secure an extra promotion. This was done, and the
results fully justified the recommendation. In all,
B. has skipped four half-grades and still continues to
secure perfect marks.
No. 12. Boy: L. M. Brother of No. 13. Under-
estimated by parents. Morally superior.
Age 6-8|; mental age 9-5; I Q 140; first grade.
Age 9-6; mental age 15-1; I Q 159; fifth grade.
At 9-6 L. passed the code and box tests of Average
Adult; also repeated 8 digits and did the ingenuity
problem in Superior Adult.
There are five children in this family, all above
average. One earned "A" marks all through high
school and graduated at 17, winning a college scholar-
ship. Father a minister of exceptional ability.
Parents 9 notes. Average parent rating on traits,
3.21. Health good. First showed unusual ability in
arithmetic at the age of 3 years. Has been allowed
to go his own pace, except as older sister taught him
in playing school. L. has a way of making for what
he wants regardless of obstacles.
FORTY-ONE SUPERIOR CHILDREN 213
One year later, age 10J. Not robust; out of school
because of unsatisfactory general health. " Brain and
ambition out of proportion to strength, but is improv-
ing. His interests take in the whole world; prohi-
bition, Bed Cross, Y.M.C.A., Boy Scouts, Athletics.
Gives morality talks to any one he thinks in need of
them. Walks miles distributing literature for all the
Has a circulating library of about fifty volumes in
constant use among the neighbor children, for which
he keeps the accounts carefully and systematically.
Remarkable in his choice of books; "has never even
by chance brought home from the public library an
undesirable book." School marks continue good.
School work rather laborious, as he has little patience
with details and makes careless mistakes. Average
rating on traits now, 8.20, considerably higher than
Teacher's notes. . All-round ability. Without self-
consciousness and speaks well before the class.
No. 13. Girl: C. M. Sister of No. 12. Early in-
dications of superior ability.
Age 7-6; mental age 11-10; I Q 158; fifth grade.
Passed the box test in Average Adult.
Parents 9 notes. Average parent rating 1.95. Health
perfect. Has abnormal physical strength. "Before
we knew it> soon after her sixth birthday, she read and
214 INTELLIGENCE OF SCHOOL CHILDREN
enjoyed the Courtship of Miles Standish, saying she
thought it was ' such beautiful language. 9 " About
the same time she wrote little two-page stories. In-
telligence was also evident in clearness of answers in
conversation, fine reasoning powers and right conclu-
sions. Superiority noted at 4 years. Allowed to go
her own pace. No home instruction except what she
received from a very bright older sister who played
school and gave her good instruction in drawing,
reading, and numbers. Reads good poetry, the Bi-
ble, and classics, all of which she thoroughly under-
stands and enjoys.
One year later, age 8|. Robust health. Highest
school marks. Leadership marked. Average rating
No. 14. Boy: J. C. A case of exceptional all-
round mental precocity. A leader.
Age 11-4; mental age 17-9; I Q 156; seventh grade.
Although only a little more than 11 years old, J.'s
vocabulary score was 75. He passed all but the paper-
cutting and ingenuity tests in the Superior Adult
group. As a result of the test he was promoted to the
Mother's notes. 3. could talk before he was a year
old, could stand at seven months and run at ten
months. He read Ivanhoe at the age of 7. "Has
seemed always to read and study. Has always been
FORTY-ONE SUPERIOR CHILDREN 215
a big boy to me." Of quiet disposition, without a
touch of vanity. Has the dictionary habit, and is an
omnivorous reader. Spends much time with encyclo-
pedias. Excellent health. Has become interested
in athletics and other boyish matters. Masters his
school work with apparent ease. Adapts himself to
any person or crowd. Leads in educational games
and is often chosen as leader. Even temper, sympa-
thetic, considerate, generous, and kind hearted. Ex-
pects to go to college and take up scientific agriculture.
Average parent rating 1.65.
No. 15. Boy: 6. 0. Illustrating all-round superi-
ority and marked precocity.
Age 12-0; mental age 16-8; I Q 189; eighth grade.
Teacher's notes. " A problem is never given that G.
will not try. He always wants to know why, and will
stick to his view until it is proved incorrect/* Has
ambition to succeed and be the first in his class. Sense
of humor far beyond his years. Has a splendid com-
mand of language. His ability is general. Has some
trouble getting on with the large boys because he still
has childish ideas about some things; but takes things
good naturedly and goes right on. "He is one of the
best all-round superior children I have ever had in
eighteen years of experience as a teacher." Average
teacher rating on traits, 2.10.
Parents' notes. Health good; sleeps ten hours.
216 INTELLIGENCE OF SCHOOL CHILDREN
Beads Shakespeare, the Book of Knowledge, and na-
ture books. Some musical ability. Specially fond
of arithmetic and history. Memory exceptionally
good. ** At 8 years he loved to be read to, and could
quote many nursery rhymes. At 4 he quoted several
long stories word for word. At 5 he could print the
alphabet, and insisted on being told how to spell
words." No formal instruction before going to school.
Since then he has gone his own pace. His questions
have been answered clearly, and current events have
been discussed in his presence. Is much interested in
machinery. Understands fairly well motor-car con-
One year later, age 18. Did not attend school last
year because of an enlarged cervical gland which
necessitated an operation. Health is good now.
Greater interest in outside activities and athletics.
Plays piano and cornet. Does his school work with
ease. School adaptability improved. Is less nervous
than formerly. Average parent rating, 2.10.
No. 16. Girl: C. 0. fflustrating marked leader-
ship and social adaptability.
Age 18-9; mental age 19-1; I Q 189; fourth year
of high school.
C. lives in a city in the northern part of California.
When we visited there and inquired for the bright-
est pupil in the city schools, the superintendent and
FORTY-ONE SUPERIOR CHILDREN 317
teachers mentioned C. without hesitation. Arrange-
ments were made with the mother for testing her the
following day. On the next day, however, C. had
come down with an attack of measles and had a high
fever. Notwithstanding this she wanted to go on with
the test, which was given, with the result noted above.
All the tests in the Stanford-Binet were passed with
one exception, the box test in Average Adult.
C. taught herself to read when she was 31 months
old. She started to school at the age of 6 9 and in
seven and a half years had completed the work of
twelve grades. Throughout she has led her classes.
She is also a leader in all kinds of school activities, such
as dramatics and class activities. She is a favorite
both with fellow pupils and teachers. Physically she
is more than ordinarily mature for her age. Her
health has always been perfect. All her extra promo-
tions have been given on the initiative of her teachers,
the parents having always urged them to hold her
back. Expects to be a lawyer.
C.'s sister graduated from university at 21 and was
president of the student body. Later did post-gradu-
ate work. Another sister of 11J years is in the low-
eighth grade. A brother graduated from university
at 0, and at 24 is holding a responsible business
position. It is doubtless this high standard of
ability in the home which accounts for the average
parent rating on traits of 2.45, or only a little above
218 INTELLIGENCE OF SCHOOL CHILDREN
No. 17. Girl: K. C. Exceptional personal charm.
Indications of musical ability. A social favorite.
Age 3-8; mental age 4-8; I Q 144; not in school.
Age 5-2; mental age 7-4; I Q 142; not in school.
Age 6-4; mental age 8-10; I Q 140; first grade.
At the age of 5 K. counted backward from 20 to 1,
gave definitions superior to use, and arranged the five
weights. This test was given as a demonstration test
before a dozen university students. K. liked the test
so well that when it was over she did not want to leave.
One of the most charming little girls we have ever
known. Absolutely unspoiled and lacking in any
appearance of self-consciousness. She sang beauti-
fully at the age of 3. Learned to name the colors, the
days of the week and the months of the year on her
own initiative and simply by asking questions.
Her father is a college professor. Both father and
mother have several relatives of superior ability.
Parents 9 notes. Nothing unusual in early child-
hood except that her development has been some-
what rapid. She spoke a few words at ten months.
Play life and social relations perfectly normal. A
favorite and takes the lead in play. Exceptional
musical ability and interest in colors. Alive to every-
thing around her. Seems to want to know everything
she hears talked about. Allowed to go her own pace,
but information she asks for is never withheld. No
formal instruction. Average parent rating, 2.00.
FORTY-ONE SUPERIOR CHILDREN 219
A year later K.'s health remains perfect, her inter-
ests are broadening and she is developing marked
traits of leadership. Average parent rating at this
M>.18. Boy:S.D. Splendid heredity. All-round
ability and exceptional courage.
Age 7-^5; mental age 10-10; I Q 146; third grade.
Second test: age 10-0; mental age 15-1; I Q 151;
The great-grandfather of S. was a chum pf Abraham
Lincoln and a candidate for United States Senator
when he died, at the age of 35. The brother of this
relative was a noted attorney. Father of S. is also
an attorney, the mother a high-school teacher. A
cousin on the mother's side is in the third year of high
school at the age of 13 years. Several uncles of the
mother were political leaders in the early history of
Teacher's notes. " I cannot say that S. has unusual
talent of any special kind; he simply has a big mind in
a big body. Ability is all-round superior." Average
teacher rating on traits, 1.5, one of the highest ratings
we have found.
Parents 9 notes. Health has always been perfect.
Spoke a few words at 6 months. Entered the third
grade at v years, and in two and a half years covered
four and a half grades. Learned to read at the age
220 INTELLIGENCE OF SCHOOL CHILDREN
of 5. Does no regular study at home, but reads about
half an hour each evening. Holidays spent in play,
chores, fishing, and swimming. No formal instruction
in childhood, but has been encouraged to stand at the
head of his class. Average parent rating on traits,
One year later, age 11-8. In the eighth grade, doing
excellent work. Health good and development satis-
factory in every way. Average parent rating at this
time, 2.10. In rating courage the mother made the
following remark: " All I can say about this is that S.
when only 10 years old entered a burning house and
brought out a baby, then reentered and dragged out a
wooden chest, and was ready to enter again when I
had to hold hi outside by force while the roof fell
No. 19. Boy: R. V. Early evidence of superiority.
Natural interest in teaching.
Age 11-7; mental age 16-6; I Q 142; high-seventh
Father a carpenter, with only a common school
education. The mother a teacher before marriage.
There are seven children, all of whom are superior.
Parents 9 notes. Health good except for an attack of
acute rheumatism when he was 6 years old. Taught
himself to read with the aid of a telephone book and
calendar. Loves to teach. Has prepared several
FORTY-ONE SUPERIOR CHILDREN 321
children in the neighborhood for school. Superiority
first evident at the age of 4. No instruction in child-
hood. " We wanted him to be outdoors and build up
a good constitution." B. is quite up to the times in
politics and war. Joins in the discussions on these
topics. Signed the prohibition pledge at Sunday
School and will not eat anything that has brandy in
it. Ambition to write books. Average parent rating
on traits, 1.90.
One year later, age 12j. Health good. School work
very good. Leads among the boys in the neighbor-
hood. Rather impatient and quick to anger, but soon
recovers his poise. Parent rating at this time, 2.00.
Teacher. Average rating on traits, 1.59.
No. 20. Boy: F. H. One of our brightest chil-
dren. All-round ability and very exceptional vocab-
Age 10-5; mental age 17-11; I Q 172; high-fifth
Vocabulary score was 78 (14,000 words). This is
almost equal to that of the average college student.
Every test was passed in year 14, four out of the six in
Average Adult, and five out of the six in Superior
Adult. Every fable was perfectly interpreted.
Father a physician. Mother had only a common
school education. Several superior relatives, a brother
testing at 137.
m* INTELLIGENCE OP SCHOOL CHDLDBEN
Parents 9 notes. Entered the first grade at 5. Abil-
ity general. Superiority first evidenced at the age of 6
by his unusual interest in school work and by his orig-
inal thinking. Has never been specially stimulated.
Allowed to go his own pace "because that was fast
enough." Average rating by parents, 1.75.
The teacher describes P. as having wonderful all-
round ability and gives him an average rating of 1.80.
No. 21 and No. 22. J. J. and B. J. Italian chil-
dren, brother and sister.
Boy: J. Age 9-0; mental age 12-7; I Q 140.
GirhB. Age 6-8; mental age 10-1; I Q 151.
Here are two Italian children, the only ones of this
nationality we have discovered testing anything like
this high. Both are exceptionally attractive, polished
yet natural in manners, beautiful and unspoiled. J. is
described as more studiously inclined than B. and as
being also more sensitive. Mother was inclined to
believe the boy the brighter of the two, but the test
places the girl slightly above. Both parents are well
Three of the four grandparents are described in such
terms as " extremely bright," "keen reader," "inter-
ested in history and international affairs," etc. The
paternal grandfather was an " able linguist and scien-
tist," a member of the Royal Geographic Society, and
FORTY-ONE SUPERIOR CHILDREN 223
a talented singer. Many relatives of culture and
learning on both sides.
J. did not learn to talk until he was 2 years old. He
is somewhat emotional, but general health is good.
Sleeps 11} hours. Learned to read at the age of 5
years. At this age about one hour daily was given to
instruction in reading and writing. " We never forced
him, but always let him know there is a premium on
fine scholarship." At the age of 6 was tutored about
2 J hours daily. Has never attended school. Specially
talented in music. Plays well and has a keen sense of
B. learned to read at the age of 5$, and was able to
read the fourth reader at the age of 6-2. " Superiority
shown in her keen observation and in her understand-
ing of human character. This was noticeable at the
age of 4, or even younger." Like her brother, has
been allowed to go her own pace.
-No. 23. Girl: M. S. A typical iUustration of the
ease with which superior children learn without in-
Age &-3; mental age 12-1; I Q 146; fifth grade.
Passed the fable test and repeated six digits back-
wards in Average Adult.
Mother's notes. M. learned to read without any in-
struction at the age of 3 years. Read signs and adver-
tisements and names on food packages which were
frequently seen about the house. At 6 years read bet-
224 INTELLIGENCE OF SCHOOL CHILDBEN
ter and more naturally than since listening to other
children at school. Has dramatic ability. Shows re-
markable grasp of all instruction and is good in execu-
tion. Makes progress two or three times as rapidly as
ordinary children. " M. has just * growed up ' as I have
had continual illness in the home and have been un-
able to give her the attention she should have had. I
held her back from skipping the fifth grade because
I felt that physical perfection was the first consid-
eration." Health always good. Wishes to become
a teacher or to take up dramatics. Mother believes
she could also succeed in business.
Later, age 9$, in sixth grade. Health and school
work A-l. "A born leader, but a little too dogmatic
and positive to be socially popular."
No. 24. Girl: M. S. General ability, combined
with talent in art. Exceptional heredity.
Age 9-1; mental age 12-10; I Q 141; low-fifth grade.
Passed the fable test in Average Adult, and the
eight digits in Superior Adult.
One grandfather a banker, the other a railway offi-
cial; both educated, intelligent men. Both grand-
mothers described as well educated and very keen. On
the mother's side James McNeill Whistler, the noted
artist, was a cousin of the child's grandfather. Several
other relatives on this side had exceptional mental
ability and physical endowment. On the father's side,
FORTY-ONE SUPEBIOB CHILDREN 225
an uncle gifted as a sculptor and painter. Several very
Parents 9 notes. Parents rated M. 1 on every trait
except courage and intellectual modesty, which they
rated 2. Physical condition has always been perfect.
Observant, excellent memory, craving for knowledge.
" Has great enthusiasm for beautiful scenery, sunsets,
and other beauties of nature." Is fond of animals. Su-
periority noted at the age of 4. Encouraged to go
ahead in school but not forced. Has been praised for
good report cards. No formal instruction whatever at
home. Ambitious in everything she attempts. Wants
to be a teacher.
One year later, age 10-*. The mother writes:
" Health good. She awakens more and more to beauty,
takes great pride in her work, and shows great love for
reading. All of her work a pleasure except arithmetic
(I wish arithmetic were a little more practical). Makes
friends easily and is very companionable with older
children. Wants to draw and loves scenery and pic-
tures. Her best chum is a school girl of fifteen years/'
Average parent rating at this time, 1.60.
Teacher 9 s notes. Unusual ability to carry a melody
in two-part singing. Reads music well. Exception-
ally good in penmanship. Superiority general. The
teacher rated all the traits 1 except general health.
No. 25. Boy: A. W. Brother of No. 26. Underes-
timated by teacher and dislikes school. Very sensitive.
226 INTELHGENCE OF SCHOOL CHTLDBEN
Age 13-1; mental age 18-6; I Q 141; low-seventh
A.'s vocabulary score was 84, which is equal to that
of the average Stanford University senior. Missed
only two tests in the scale, the ingenuity test and re-
peating seven digits backwards.
Both A. and his sister are very superior but A. seems
to be more original and better informed. Until a few
months before the test A. had always attended a coun-
try school. His grades in school are good, but not ex-
ceptionally superior. He has no hesitation in saying
that he does not particularly like school. The teacher
rated him 3 ("average ") on all but two of the twenty
traits. She sees nothing exceptional in this boy's men-
tality, although he is better informed and has a larger
command of language than the average teacher.
One wonders whether the teacher's misunderstanding
has anything to do with the boy's dislike of school.
Parents 9 notes. Health good except for chorea,
which has now practically disappeared. As a small
child he was very timid, and he is still sensitive. Re-
markable memory, which first showed itself at the age
of 4, when he learned his story books by heart. At
that age he also learned most of Foe's The Bells. Has
always used big words correctly. Learned to read at
the age of 6 $. In three or four months he could read
all of Riley 's child rhymes. " Prom the time when he
was' a young child A. has seemed to have understand-
FORTY-ONE SUPERIOR CHILDREN 227
ing and knowledge in almost everything beyond his
years." Draws exceptionally well and has mechanical
ability. "At four years could repeat verbatim pages
and pages of books which were read to him." Allowed
to go his own pace because of his tendency to nervous-
ness. The only instruction has been in the form of
answering innumerable questions. Several relatives
of very superior ability. Average parent rating on
Later, age 14-2. Health good, school marks im-
proved; school work easier; less nervous.
No. 26. Girl:E.W. Sister of No. 25.
Age 11-5; mental age 16-11; I Q 148; high-seventh
All the tests in Average Adult passed except the
code; eight digits direct order, and seven digits re-
versed passed in Superior Adult.
Parents 9 notes. Age of talking, 20 months. Health
excellent. Has always been intellectually alert beyond
her years. Ambitious to excel. Is very practical
Has always had an excellent memory and early learned
nursery rhymes and jingles. Superiority first noticed
at the age of 4. Is musical. Allowed to go her own
pace, " as she seems inclined to go quite as fast as is
good for her." No formal instruction at home. De-
sires to become a teacher*
228 INTELLIGENCE OF SCHOOL CHILDREN
No. 27. Boy: R. K. Exceptional heredity.
Age 8-9; mental age 12-4; I Q 141; fourth grade.
Age 11-4; mental age 16-8; IQ147; high-seventh
Father a mining engineer, mother a teacher. Pa-
ternal grandfather a teacher of superior ability. One
uncle a doctor of divinity and " a bright scholar." One
cousin is a "mechanical engineer of exceptional abil-
ity." Another cousin, a post-graduate of Harvard, is
said to be one of the best mathematicians that Harv-
ard has had in years. Relatives farther back on this
side were Roger Williams and Colonel Crawford.
Maternal grandfather a teacher and lawyer of abil-
ity; maternal grandmother a teacher and "a great
student up to the age of eighty years." Two uncles
and one aunt on this side had exceptional mental
ability. One cousin is an artist of ability, and an-
other a talented singer. Washington Irving was a
cousin of the great grandfather. Another noted rel-
ative farther back was an earl of Kilnockie.
Parents 9 nates. R. is somewhat nervous; otherwise
health is perfect. Nothing unusual in early life. En-
tered the second grade at 6 years, and later skipped
half of the fourth and half of the sixth. Never urged
on. Best work is in English and music. In his com-
positions shows unusual appreciation of language. Is
ambitious to write. Average parent rating on traits,
FORTY-ONE SUPERIOR CHILDREN 229
One year later, age 12-6. " R. is finishing the eighth
grade with excellent marks. Work very easy for him.
Health good." Average parent rating at this time,
No. 28. Boy: J. P. Underestimated by parents.
An exceptionally logical mind.
Age 8-1; mental age 10-10; I Q 134; third grade.
Age 9-2; mental age 13-0; I Q 141; fifth grade.
Age 11-4; mental age 16-6; I Q 137; seventh grade.
Parents 9 notes. The father, a college professor, was
slow to believe that J. was much above the average
child in ability. He has no brothers or sisters, and the
parents had no general standard by which to judge
him. Average parent rating, 2.44. Nothing unusual
in early life, health, or training. Was taught to read
at the age of 6, but has had no formal instruction.
Two years later, age 11. Health good. Tonsils
recently removed. School work done without effort.
Somewhat nervous and sensitive. Average parent rat*
ing at this time, 2.20, or somewhat higher than before.
Teacher's notes. " J. can stagger you with astro-
nomical facts. Delights in historical stories. Is not
contented with statements made in the text, but wants
detailed information. Questions everything; loves
an argument and debates with zeal and ability. Was
wildly happy when appointed to lead a debate. Has
a code and loves secrets. His mind is alert to every
230 INTELLIGENCE OF SCHOOL CHILDBEN
impression. His hands are not responsive; he dis-
likes to write or draw, but grits his teeth and does
average work to avoid having to do it over. Reason-
ing is his strong point. He can read any book and
repeat the substance of it months afterwards. Social
adaptability normal, but rather prefers to play alone.
Does not care for conventionalities. Has an unusual
sense of justice." Average teacher rating, 2.00.
No. 29. Boy: B. H. Very much underrated by
Age 9-7; mental age 13-10; I Q 144; low-fifth grade.
The interesting thing about this child is that the
teacher considers his ability " average, except in lan-
guage." As a matter of fact, he is farther advanced in
vocabulary than in his general mental development.
He is under-age for his grade, and has been rated by
the teacher in comparison with children two and three
Fortunate heredity. Two uncles on the mother's side
unusually intelligent; one was a prominent lawyer
when he died at the age of 35, the other entered high
school at 11 and is now editor of a large city news-
paper. A great-great-uncle of the boy was a doctor
of divinity and one of the foremost of pulpit orators
in the South. B. has two brothers almost as bright
Parents 9 notes. Health perfect, but sleep not very
FORTY-ONE SUPERIOR CHILDREN 231
sound. " Have encouraged him because he is not am-
bitious/' Was given no home instruction except for
about a year when he started to school. Average rat-
ing on traits, 2.65.
One year later, age 10J. School work good, except
that his penmanship is poor and his written work
somewhat untidy. This sometimes lowers his grades.
Improvement in social adaptability. Average parent
rating at this time, 2.45.
No. 30. Boy: L. 0. Underrated by parents. An
"only " child. Marked precocity.
Age 8-3; mental age 12-2; I Q 142; high-fourth
Passed the clock test and the induction test in 14,
and in Average Adult repeated six digits backwards.
Vocabulary score, however, only 25, which is not more
than a year above his actual age.
The most interesting thing in the data furnished by
the parents is the fact that they rate the child 3, or
average, on eighteen out of twenty traits. One won-
ders whether this is because L. is an only child and
there is no standard of comparison in the home. The
average teacher rating is 2.20.
Parents 9 notes. Health good. Learned to read at the
age of 4. Above average in power of concentration.
Became interested in books at the age of 2 years.
Was persistent in effort to understand meaning of
382 INTELLIGENCE OF SCHOOL CHILDREN
words and characters. Memorized and recited readily
at the age of 2 years. "Have not encouraged him
to forge ahead, but have not held him back. Little in-
struction at home beyond the teaching of sounds of let-
ters. Have also tried to answer all his questions and
to point the way to further investigation. We think
the child has a tendency to read too much."
No. 81. Boy: C. M. Underrated by teacher. Ex-
Age 8-6; mental age 12-0; I Q 141; high-fourth
This case is mentioned chiefly because of the follow-
ing statement of his teacher: " I would say, taking my
class as a basis of judgment, that C. is an average
child." The teacher admits that he is doing excellent
work in the high-fourth grade, also that the class is an
unusually satisfactory one. She neglects to note that
the average age in her class is about 10 years, while
that of C. is only 8J.
A sister of 15 is almost as bright as C. The father is
a minister, a graduate of a theological seminary. Ma-
ternal grandfather an army officer and graduate of
Oxford. Maternal grandmother very musical. Of
eight uncles, two were able lawyers, and three were
successful engineers. The mother's grandfather was
one of the most prominent Canadian statesmen of
his day. Paternal grandfather a college graduate;
FORTY-ONE SUPEBIOE CHILDEEN . 233
paternal grandmother musical. The only uncle on
this side is an expert chemist, whose sons show unus-
ual ability in literary lines. The father's grandfather
was one of the leading spirits in the old Hudson Bay
No. 82. Girl: M. C. Brightest girl in the Stan-
ford records. Superior family of children, ordinary
Age 7-10; mental age 13-8; I Q 174; fifth grade.
This child, tested by Miss Blanche Cummings, Di-
rector of Special Classes in Fresno, California, is the
brightest girl of whom we have a record at Stanford
University. Her development will be carefully fol-
The father is a jeweler; the mother was a milliner
before marriage. Neither parent had more than a com-
mon school education. There are three other remarka-
ble children in the family; a sister, age 11, in the sev-
enth grade; a brother, age 10, in the fifth grade; and a
brother, age 6, in the second grade. The last named
tested at 136. No other relatives of superior ability
are known to the parents.
Parents 9 notes. Nothing unusual in health or physi-
cal development in early childhood. Was given no in-
struction, but learned to read by her own efforts at
three years. Was permitted to use a typewriter and
with it learned her letters, figures, reading, and spell-
234 INTELLIGENCE OF SCHOOL CHILDREN
ing. Entered school only a year and a half ago, and
has attended four different schools. Notwithstand-
ing these frequent changes she has made five grades in
that time with ease. " M. is more inclined to be firm
and stubborn than the other children." Desires to be*
come a school teacher. Average parent rating, $.70.
No. 33. Boy: P. E. Early indications of superior
intelligence. Apparent moral inferiority.
Age 9-5; mental age 13-9; I Q 146; seventh grade.
Age 11-3; mental age 16-10; I Q 150; first year high
Both parents physicians. Maternal grandfather a
journalist and politician who knew seven languages.
A cousin on the mother's side is said to be as bright
as P. Mother's relatives chiefly doctors, lawyers, and
Parents 9 notes. P. knew his letters at 14 months and
could read at years. Learned to count at the age of 2
years. A little later knew numbers as far as the thou-
sands and could find numbers in the telephone book.
"When 4 years and 3 months old had read a good part
of the Bible and read as well as a boy of thirteen." Al-
ways insisted until told what he wanted to know. En-
tered the third grade when he started to school at 6,
and made nine grades in four and a half years. Memo-
rizes very rapidly. Once became possessed of a desire to
know the location of every town, river, and mountain
FORTY-ONE SUPERIOR CHILDREN 235
and read the atlas through. Great collector of stamps,
coins, foreign transfers, etc. " I have encouraged him.
I consider a child may just as well be learning some-
thing as to fool away his time. I never made him
study. While he was a small child I bought blocks
with letters and numbers, maps to be put together,
geographical games, alphabetical and numerical boards
and other playthings with which to learn. Later
bought him a typewriter, which he soon learned to
use." Play life fairly normal, but made difficult by
the fact that his classmates are much older and larger.
With them he cannot be a leader, while with younger
children he is somewhat domineering and bossy.
Obeys while at school, but is rather selfish and im-
perious at home. The mother accounts for this by
the fact that he is an only child and has been allowed
his own way. Mother's average rating on traits,
3.70. Desires to become a professor of mathematics
Later, age Id. Was out of school last term and
worked as collector for a newspaper, making $25 a
month. His vocational ambition now is to be a
banker. School marks still excellent, but hardly as
good as before. " Still somewhat spoiled, selfish and
occasionally unkind in his criticisms of others. Com-
pels boys of his size to do as he says. Strong willed.
Punishes himself rather than give in. Needs a man's
Notes from school principal. Undoubtedly great
286 INTELLIGENCE OF SCHOOL CHILDBEN
native intelligence. Seems very learned. His ability
is certainly general. Often appears not to be atten-
tive, but later surprises one by being able to repeat
eveiything that has been said. However, is spoiled
and vain and is looked upon with a certain amount
of distrust. Is said to have abnormal sex inter-
ests. Once attacked a small boy with a knife. Effu-
sively affectionate toward his teacher, but disrespect-
ful toward his parents. Stubborn and willful. His
school conduct, however, absolutely beyond reproach.
Teacher's notes. Ability rather one-sided. Remark-
able memory for facts, but lack of judgment. Has
few playmates. Reputed to be a bully among
younger children, although he did not show this at
school. Is tyrannical toward his mother and grand-
mother. Average rating, 2.89.
Another teacher states: "His analysis, interpreta-
tion and memory for detail in Julius Ccesar and Ivan-
hoe have been far above the average of his class."
This teacher described P.'s ability as general rather
than special, and gave an average rating on traits
There is no question about this boy's unusual abil-
ity. Some would perhaps account for it on the
ground of his early instruction, but we doubt the val-
idity of such an explanation. The boy's social and
moral development does not promise well, although
his present objectionable tendencies may be out*
FORTY-ONE SUPERIOR CHILDREN 37
grown later. This is our only superior who has evi-
denced questionable moral traits.
No. 34. Boy: H. H. Early instruction accompa-
nied by marked indications of superior intelligence.
Age 6-0; mental age 9-4; I Q 156.
Age 8-9; mental age 12-10; I Q 147; seventh grade.
In the second test, age 8-9, the vocabulary score
was 55 (nearly 10,000 words). This is better than
the median for 14 years.
Little is known of the ancestors of H. except that
both of his grandfathers were farmers with only a
common school education. One distant relative was
a lawyer of national reputation. The father is a
teacher and the mother a woman of marked intellec-
tuality. The accomplishments of H. were exploited
in a number of newspapers in 1918-13.
Parents 9 notes. Was specially instructed in early
childhood by the mother, who early began reading to
him such literature as Hiawatha, Julius C&sar, Bible
Stories, etc. Learned to read at 4. At 6 was able to
add, subtract, multiply and divide numbers as far as
the millions, to keep the family accounts, make up
bilfe, etc. Mastered the number combinations by
playing dominoes, and learned a great deal of geog-
raphy by playing post office and writing addresses on
envelopes which he gave to his mother. Has accum-
ulated a rich store of knowledge about nature. All
238 INTELLIGENCE OF SCHOOL CHILDREN
his work is play to him. Plays the piano quite re-
markably for a child of his age.
Later, age 8. Not specially fond of school. Does
little home study; reads only three or four hours a
week. Spends most of his time at play. Mother's
rating on traits, .10.
Teacher's notes. A very lovable child, and below
average only in leadership and initiative. " Wonder-
ful knowledge of history. Is always ready with sto-
ries to illustrate a point. Especially good in oral
composition. Large fund of general information."
Ability slightly one-sided. Does not seem to fit in
with the play life of his classmates. Beads a great
deal, including Dickens's novels, Shakespeare stories
and child verses. Rather restless. " Exceptionally
poor in writing and other handwork, but amazes one
with his knowledge of historical events." Average
rating on traits, 1.95.
Later, age 9-10. Described by the succeeding
teacher as in good health, growing very rapidly and
more interested in play and companions. Marks
still high. Improvement now in social adaptability.
Average rating on traits at this time, 2.10. " When
he came to us a little over a year ago he was ex-
tremely restless and timid; spoke in very low tones,
flushed easily, and never volunteered remarks. Re-
cently he has relaxed, plays ball, worships the big
boys, and has sprouted physically. With this has
come a sudden interest in assigned tasks which seems
very promising for the future."
FORTY-ONE SUPERIOR CHILDREN 239
No. 35. GirlE.W. Ill-health. All-round abil-
ity. Exceptional heredity.
Age 14-8; mental age 19; I Q 134; high-seventh
E. is in the grade corresponding to chronological
age. However, she was out for two years at one time
and has missed at other times on account of illness.
Although she has attended school only intermittently
for five years, in this time she has completed seven
grades. Considering her health, it is perhaps best that
she has not been promoted more rapidly.
The teacher says : " No matter what E. has to do it is
always well done. She has a wonderful power of con-
centration, a keen sense of humor, and never gives up
until the battle is won. Her examination papers are a
wonder to her teachers. They are always to the point
and definite. E. has decidedly all-round ability.' 5
The teacher rated her 1 on every trait except general
Parents 9 notes. A good deal of ill-health from three to
ten years. Somewhat nervous and irritable when fa-
tigued. "Reads as many hours as we allow." Has
spent many happy hours delving into children's ency-
clopedias. Never had any formal instruction at home.
One sister and one brother of little if any more than
average ability. ** E. has always been first to grasp the
meaning of a game, puzzle, or any subject under dis-
cussion. 9 ' Average parent rating on traits, 1.65.
240 INTELLIGENCE OF SCHOOL CHILDREN
One year later, age 15-4. Health not quite so good.
Marks satisfactory but not quite as high as formerly.
Inclined to worry over her school work. Average rat-
ing now, 1.95.
The maternal grandfather was a school teacher of
" fine ability.*' Maternal grandmother " a student to
the age of 82." Uncles are successful professional and
business men. One aunt a talented musician. The
mother's brother was a leader in his university class,
but became insane. The paternal grandfather and
grandmother were school teachers. One uncle on this
side is a lawyer and judge. E.'s great grandfather
served the longest term in the New York state legisla-
ture of any man up to his time. Of two other relatives
on this side, one was a noted Congregational minister
in New York City, the other a famous surgeon.
No, 86. Soy: J. E. Exceptional heredity. Diffi-
culty in social adjustments.
Age 11-0; mental age 16-1; I Q 146. School work
irregular, but chiefly in the fifth and sixth grades.
J. made the remarkable vocabulary record of 74
correct definitions. He also passed the test of repeat-
ing eight digits in Superior Adult.
Has been kept back in his studies by ill-health (in-
cipient kidney trouble), from which he had largely re-
covered at the age of 15. Has been educated by a gov-
erness and in private schools. Teachers consider him
FORTY-ONE SUPEETOR CHILDREN 241
very unusual in ability but hardly up to average in
J.'s greatest difficulty has been in submitting to for-
mal instruction and in adapting himself to other chil-
dren. Until 11 years old he had but few opportunities
to associate with others and was considered more or
less erratic. His social adaptability, however, has
steadily improved, as has also his tendency toward ir-
ritability and imperiousness.
Both of J.'s parents are of English descent. The
father is a scientist, educator, and publicist. Two of
J.'s brothers are of average mentality; one sister, now
dead, was very superior, and another sister is a woman
of very exceptional ability. A number of superior
relatives on both sides. One uncle, on the mother's
side, was an Admiral in the United States Navy.
Relatives farther back distinguished. Many distin-
guished relatives on father's side, one of whom was
Ralph Waldo Emerson.
Parents 9 notes. As an infant J. was much disturbed
by loud or sharp noises. Showed superior ability
early. While still a small child drew diagrams of
"inventions" which proved to be actual parts of
machines he had never seen. Great interest also in
astronomy. Listed stars of the fourth magnitude. At
present (age 11) works in the laboratory with shells,
doing a grade of work which few university seniors
can surpass. Will soon publish a book on California
shells. Expects to become a scientist.
24.2 INTELLIGENCE OF SCHOOL CHILDREN
Four years later, age 15. Health improved. Some
lessening of enthusiasm in scientific work on shells,
accompanied by marked increase of interest in manly
sports. The expected book has not been finished.
Excellent school marks; work done with decided ease.
Social adaptability now average. An awakening sense
of responsibility. Average parent rating, 2.40. At
this time J.'s teachers also testify to his marked im-
provement along social lines,
No. 87. Boy: M. A. Inferior school work and
marked lack of social adaptability.
Tested first at 10-11; mental age 15-0; I Q 137;
When tested nearly a year later the I Q was 138 and
he was in the first year of high school.
Heredity exceptionally good. Father an able law-
yer; mother formerly a teacher in a city normal college.
Many prominent men and women among his relatives,
one of whom was Samuel Adams.
An exceptionally bright boy but a problem for his
teachers. Although his mental age is well above the
average in the first year of high school, his grades run
from C to D. Is temperamental and more or less
queer. Easily takes a dislike to teachers or classmates.
Regards his school work with more or less contempt
and part of it he refuses to try at all. Because his
school work is poor some of his teachers consider his
intelligence only average.
FORTY-ONE SUPERIOR CHILDREN 243
Teacher's notes. An unusual ability to associate
facts, particularly scientific facts, and to repeat from
memory after one reading. Rated 4 in social adapta-
bility, leadership, emotional self-control and unselfish-
ness. Is babyish in his play. Nervous; has muscular
twitchings and is easily embarrassed in class. Is se-
lected by his fellows as the one to tease, torment, and
nickname. Cries easily. "However, I believe that
M. will become more adjusted to his surroundings and
make a superior man." Average teacher rating on
Parents' notes. Health good. No special instruc-
tion in childhood except the little he received from a
workman on the ranch. Has been held back, but is
now allowed to go his own pace. Reads history, scien-
tific works and all kinds of magazines. Desires to
become an inventor. Average parent rating, 1.80.
(This was one of the few cases in which the parent's
ratings averaged higher than those of the teacher.)
Later, age 13. Marks in high school now slightly
above average, and there is marked improvement in
social adaptability and emotional life. Average parent
rating at this time, 1.60.
No. 38. Boy: A. L. 8. Poetic talent combined
with all-round ability.
Age &-4; mental age 13-2; I Q 141.
This child was first brought to our attention as a
result of a group test. We have not yet had oppor-
244 INTELLIGENCE OF SCHOOL CHILDREN
tunity to learn much about him, except that he is con-
sidered one of the brightest pupils in the school of the
small city where he lives. The following poem was
composed when he was 9 years old. It shows remark-
able maturity of thought for a child of his age:
Do not worry over trifles, though to you they may seem
All your fretting will not help you, or your troubles dissipate.
If your sky is dark and gloomy, and the sun is hid from
Bravely smile and keep on smiling,
And your friends will smile with you.
Happiness is so contagious, and a smile is never lost;
Then why worry over trifles, though your heart seems tem-
Therefore go on life's rough journey with an optimistic
See the world is good to live in, and that living is worth
No. 39. Boy: J. 8. Intensive mental culture in
early childhood. Fine mental balance. Has a sister
who is an infant prodigy.
Age 9-6; mental age 16-4; I Q 172; sixth grade.
Age 10-4; mental age 17-8; I Q 171; seventh grade.
In the first examination, age 9, J. passed four tests
in Superior Adult, including paper cutting, eight digits
direct order, seven digits reversed order, and the in-
FORTY-ONE SUPERIOR CU1LUKEJN 245
Special interest attaches to J. because he is a brother
of Martha, who at the age of 26 months was able to
read any primer. 1
Father a lawyer and a man of more than ordinary
ability. Graduated from university at 81. Mother
a teacher before marriage. Maternal grandfather a
farmer, of common school education and average abil-
ity. Uncles and aunts average or somewhat above.
Paternal grandfather a bookkeeper of business college
education and average ability. Paternal grandmother
of average ability, common school education.
Father's notes. J.'s superior ability first evident in
third year. Father accounts for the superiority as
" due to the fact that we deliberately set ourselves to
the task of educating him when he was a young child.
When J. was a mere baby I determined to start his
education. Commencing at the age of two years I
adopted artifices to make his play a source of education
and kept at it persistently until he was five years old
and had acquired the fundamentals of the first three
years of school, after which I dropped the matter. In
the case of the second boy, I had no time to take that
course and did not do so." (Second boy only average.)
Father describes J. as serious and dreamy, finding
his greatest pleasure in reading. Little interest in
tools or machinery. Quite different from the boister-
ous, happy-go-lucky younger brother. If left to his
1 See Tlic Journal of Applied Psychology, 1918, pp. 219-28: "An
Experiment in Infant Education."
246 INTELLIGENCE OP SCHOOL CHILDREN
own devices would spend all his leisure reading.
Health always perfect except for scarlet fever at 5
years. Average rating on traits, .35.
One year later, age llj. Health good; adenoids
and tonsils recently removed. Average of father's
rating on traits now 1.75. Marked improvement in
Teacher's notes. " J. is a boy of wonderful ability
for his years. In arithmetic he never draws an un-
warranted conclusion or premises anything unneces-
sary to the conclusion." When he started to school
he covered the first grade in a half day, the second
grade in two months, the third grade in six months,
and the fourth grade in two months. All but one of
the twenty traits graded 1 by the teacher, with special
emphasis on the boy's lack of vanity.
Play interests and play life described as normal. No
physical handicaps, nervousness, or eccentricities of
any kind. " In every respect normal with the excep-
tion of superior intelligence."
No. 40. Henry. Illustrating the relative inde-
pendence of I Q and schooling.
Scientific ability overshadowed by musical genius.
As a near neighbor boy, Henry has been under our
observation since the autumn of 1910. At that time
he was a little more, than 1&J years of age. He was
tested at 14$, earning the mental age of 19 (I Q 131).
FORTY-ONE SUPERIOR CHILDREN 247
Although the I Q is satisfactory, it is matched by
scores of others among our records; but there is only
Henry had never been to school except for a few
months when he was 6 years old. He lived in a little
shanty with his semi-invalid mother and was the sole
source of income for the support of her and himself.
He tramped often to the mountains in search of rare
wild flowers which he brought home and sold in
beautiful bouquets to people who knew him. Some-
times he weeded lawns or did garden work for his
neighbors. For some years also he served as janitor
for a little rural school near his home. His earnings
rarely amounted to more than $15 a month, but some-
how he and his mother managed to live on this
Henry's mother, since dead, was a woman of refine-
ment and intellectuality, the author of two novels and
a number of poems. She also wrote essays on socio-
logical questions, at least one of which was published
in an English periodical of international circulation.
She was an idealist, imbued with advanced notions
regarding religion, sociology, and woman's place in
Henry's mother was almost 50 years old when he
was born. His father was an unsuccessful member
of a distinguished family. Henry's paternal grand-
father was an Archbishop of Ireland, and dukes and
earls are numbered among his cousins.
248 INTELLIGENCE OF SCHOOL CHILDREN
Shortly after Henry started to school, at the age of
6 years, he was one day seized on his way home from
school with a strange muscular paralysis. He fell to
the ground and had to drag himself home. Chorea
set in, from which he suffered severe recurrent attacks
for years. Except for occasional twitchings, he had
fairly recovered at the age of 14, and somewhat later
his recovery was practically complete. On account of
this nervous tendency, however, his mother did not
see fit to send Ti. to school, nor did she give him much
formal instruction at home. She talked with him
endlessly, read to him occasionally, and sometimes he
read to her. They discussed religion, politics, and
matters of literature and art. We have a list of over
three hundred books which Henry had read before he
was 14 years of age, also bulky notes of extensive con-
versations which we had with him on such questions
as socialism, atheism, scientific problems, etc. At
14 he discussed these matters with greater breadth of
knowledge and much deeper understanding than the
average university senior. No less striking was his
ignorance in certain school subjects. His spelling was
wretched, and he had studied no formal arithmetic
above the four fundamentals and simple fractions.
As a boy of a dozen years, Henry's appearance was
odd and interesting in the extreme. His speech was
quaint, and rather drawled and stilted; his face was
childish, but he looked at you with eyes that seemed
utterly void of self -consciousness; his clothes were
FORTY-ONE SUPERIOR CHILDREN 249
often ragged and always ill-fitting; his hair hid his
ears and straggled down to his shoulders; his face and
shoulders twitched occasionally with choreic spasms.
Everybody considered Henry as queer, not to say
freakish. If employed to weed a lawn he was likely
to forget what he was doing while trying to compose
and whistle a tune. His janitor work was hardly more
successful. Henry had shown promising ability with
the violin at the age of five years, but his chorea had
put an end to his musical practice. Neither violin
nor piano was touched again until he was about 15
years of age. His musical talent, however, survived
all the vicissitudes of poverty and illness. Henry
knew that his nervousness, and still more the effect
of hard labor upon his hands, had ruined forever the
hope of his becoming a great musical performer; but
he would become a composer. Day and night he
dreamed of this and wrote out in musical notation
At the age of 15, having practically recovered from
his chorea, Henry resolved to gratify a long cherished
ambition he decided to purchase a piano. He found
an old second-hand one and bought it for $60.00, which
sum he managed to save out of his scanty earnings by
doing without various " necessities " of life. Although
he had not tried to play on the piano before, within a
year he was giving recitals among his university friends.
Within three or four years his playing was quite re-
markable. Shortly after this his playing was brought
250 INTELLIGENCE OF SCHOOL CHILDREN
to the attention of prominent musicians in San Fran-
cisco who, with other friends, gave him encouragement
and help. He was placed under the instruction of one
of the best music teachers in the West, and soon took
rank as one of the most promising pupils that instructor
had ever had. At the age of 19 he spent several months
in New York. His compositions at this time were
pronounced promising by various prominent musi-
cians. At the age of 0, without ever having been
in school a year in his life, Henry was made Instructor
of Harmony in the summer school of a great state
university. He was reappointed for a second year,
but was soon afterwards taken for military service. 1
Those who had considered Henry as merely a queer
child with impossible ideas and exasperating manners
and frankness, were finally compelled to admit his
musical ability. Even then, however, he was gener-
ally considered a freak in all but his musical ability.
His general intelligence has never been correctly ap-
praised by the majority of his friends.
We have seen the verdict of the Binet test. As the
result of many hours of conversation with the boy,
over a period of many months, we are convinced that
1 On the day when the author read the proof of this chapter a
letter reached him from one of Henry's friends in which was the
following statement: "Professor S (head of the Department
of music in the university referred to above) says of Henry that
he is the only American known to him who has really great talent
for musical composition. Damrosch has promised to produce his
first symphony, almost finished when he enlisted, as soon as it is
put in final shape. About a dozen of his compositions are being
published at the instance of Professor S ."
FOETY-ONE SUPEBIOR CHILDBEN 251
his ability in science was almost as great as in music.
Before the age of 12 he had read university textbooks
in botany. His knowledge of California wild flowers
at this age was remarkable. He had studied seriously
the principles of plant breeding, and for a time, when
it seemed impossible to realize his musical ambitions,
he considered botanical science for his life-work. He
might have done so but for the fact that his educa-
tion had been too irregular to permit him to enter a
One of the most noticeable things about Henry has
always been his independence of judgment. His
opinions on all kinds of matters are quite pronounced,
and he expresses them without regard for other peo-
ple's feelings. By many acquaintances he is consid-
ered rude and ill-mannered. This does him injustice;
he is merely naively honest, due both to his tempera-
ment and to the influence of his early training.
It remains to be seen whether Henry will become
one of the famous musical composers of his day.
Several musical critics of note hope for this outcome.
If he attains fame as a musician, his biographer is
almost certain to describe his musical genius as natu-
ral and inevitable, and to ignore the scientist that he
might have been.
No. 41. Boy: D. B. Indications of real genius.
Unequaled intellectual spontaneity.
Age 7-4f ; mental age 13-7; I Q 184; not in school.
252 INTELLIGENCE OF SCHOOL CfflLDEEN
This is the highest intelligence quotient we have
ever found, and all the supplementary data indicate
that there is no other child in our list who equals D.
in all-round intellectual ability.
The test was made before a class of about a hundred
students at Columbia University. The day was one
of the most uncomfortable in the history of New York
City, the official temperature for the day being above
100 degrees. The room was close, ill-ventilated and
wretchedly hot. The test began with year 9. All of
the tests of this group were passed. In year 10 all of
the tests were passed except that of drawing designs,
which fell just short of being satisfactory. In year 12,
seven of the eight tests were passed with ease. The
three disarranged sentences were given without a sin-
gle error in 12, 10, and 5 seconds. The five fables
were interpreted as follows:
1. Hercules and Wagon Driver: "H you work yourself
you will get help."
& The Milkmaid and her Plans: "Do not build castles in
3. The Fox and the Crow: "Do not listen to flattery."
4. The Farmer and the Stork: "If you keep company
with bad people you will have to suffer the conse-
5. The Miller and the Donkey: "Stick to one way."
In year 14 the induction test was passed without
error, the rule being given as follows: "You multiply
by two each time." The other tests passed in this
FORTY-ONE SUPERIOR CHILDREN 253
year were president and king and arithmetical reason-
ing. There was only one success out of three trials
in the clock test. In Average Adult the fables and
box test were passed. Although the examination cov-
ered a wide range of tests, it required only 45 minutes.
The responses were perfectly natural, almost playful.
There was no waiting for applause, no appearance
whatever of vanity.
Although D. was not enrolled in school at the time
of the test, he regularly attended the playground
activities at the Horace Mann practice school. Pre-
viously he had attended a kindergarten. All of his
teachers had recognized his phenomenal ability.
Father, Russian-Jewish; mother, Polish-Jewish.
The father is an advertising man and writer, and has
published three books of fiction. The mother is a
high-school graduate, did some work in a university,
and has written short stories and poems for various
periodicals. Maternal grandfather a business man
of " high intellectuality." Two cases of unusual mu-
sical ability on the mother's side, also several distin-
guished rabbis* Paternal grandfather a business man
of unusual mechanical ability, fond of making and
solving puzzles. The paternal grandmother taught
herself to read English late in life. Rabbis on this
D. is an " only " child. The mother is a woman
of exceptionally keen and judicial mind, and has kept
bulky notes on D.'s mental development since he was
254 INTELLIGENCE OF SCHOOL CHILDREN
a baby. She has furnished us with the following
interesting items of information:
"D. stood alone between five and six months;
walked at nine months, and talked at about a year.
First teeth between four and five months. Nursed
for only five months. No illness except measles and a
light case of chicken pox. No physical defects. Sleeps
about 11 hours.
" Played with anagrams when a baby and learned
to read as gradually and naturally as he learned to
talk. At three, without us knowing he could do it, he
picked up a new book suitable for children of nine
years and read it through intelligently. Has had
some private lessons in music and gymnastics. Has
also taken a few lessons in interpretative dancing.
" Dresses and undresses alone, bathes himself, cleans
his teeth alone and tends to his bodily needs. Plays
ball, bats and skates. Handles ' mechano ' models re-
quiring deft fingers. Typewrites rapidly, using only
two fingers on each hand. Taught himself printing
" Reads very rapidly. If he likes a book will return
again and again to it, memorizing the parts he specially
cares for. Probably averages eight or ten hours a
week reading. Leaves his book willingly to play, but
goes back to it when play is over.
"Recently a world atlas, baseball guides, and base-
ball news in the daily papers have all furnished him
with what he calls * important work/ Has read a
FORTY-ONE SUPERIOR CHILDREN 255
great deal of Shakespeare with a particular liking for
the historical plays. (Pericles is his favorite.) His
knowledge of Shakespeare characters is amazing.
Reads the Book of Knowledge and as many animal
stories as he can lay his hands on. Desires to travel in
order to see and learn the habits of wild animals. Has
read every history book in the house, including Gib-
bon and Grote. He criticized Gibbon as 'having
left too much out ' in writing about Rome. Among
his papers are sundry notes marked * Important things
the Scottish kings did/ * List of Roman Emperors and
what they ruled over/ etc. This shows that he reads
to find out things which he considers important.
When taken to the public library he invariably chooses
books of history. Is very fond of fairy tales but has
not been permitted to read many.
"D. will carry through projects extending over long
periods. It took him several days to complete a map
of the apartment drawn to scale; many weeks off and
on, to complete a geographical map of his imaginary
country, ' Borningtown/ and for a year he spent much
time recording foreign state automobiles sighted in
New York, with directions for recognizing the various
licenses. Has notebooks and papers covered with base-
ball data. Keeps data embodying special features of
maps, charts, etc. In reading Shakespeare pays care-
ful attention to the notes on the text, which in the edi-
tion he is reading (Knight) are voluminous. A recent
interest which has taken the place of the foreign autos
256 INTELLIGENCE OF SCHOOL CHILDREN
is that of the trolley system of New York City. His
pockets bulge with notes and transfers (together with
marbles, with which he plays at every opportunity).
" Plays games with cards, a baseball game, and a
question game. The latter is an information contest.
In the game of characters his side always wins, for he
has an inexhaustible supply of Shakespeare characters
to draw upon. Similarly when it comes to cities or riv-
ers, such sources as Russia furnish him a supply which
no one else can compete with. Other games which he
likes are various kinds of solitaire, chess, and quite a
difficult game shown to him by a teacher of mathe-
matics, a game in which he outplays every one by his
unerring calculation in what he called its 'double
The foregoing notes refer to D.'s reading and abili-
ties prior to August, 1917. In March, 1918, the mother
writes as follows:
"His Shakespeare interest holds, but he has read
recently much less history. Has developed an interest
in the scientific articles in the Book of Knowledge. Re-
cently showed me a toy telescope which he had made
out of his old miscroscope and mounted on the steel
parts of his mechano. Spends hours over his toy train
tracks. Once calculated how long it would take his
little train to run a mile at the rate it went around his
track; measuring in the center of the track, he ex-
plained, ' to be sure to get the exact answer.'
"Last year his expressed vocational ambition was
FORTY-ONE SUPERIOR CHILDREN 257
to be a baseball player. Later, he said that while he
h&d not given up his plan to be a baseball player, he
had decided also to be an author. This was while he
was deep in his ventures of book writing, having be-
gun three or four different books in the fall of 1917
and finished a play for his mother's birthday. He
has begun a book called Bomingtown, with chapters and
headings already planned; also another book called
Facts about Bomingtown and Washabett, with table of
contents and headings for fifty chapters ! Of the text
so far there are five typewritten pages and one illus-
tration. A third book is about Bully Wu&y, or the
"Another new interest is the dictionary he is mak-
ing of 'Borningtown/ Many of the words which he
makes up for this dictionary are intended as improve-
ments on the English language. For example
'smallen,' to make small. His interest in words and
their derivations led us to begin this year a little
formal Latin, at which he spends about an hour a
week. His ability to analyze and classify have made
it quite easy for him to learn thus far the first and
second declensions of nouns and adjectives and a few
"He learned to count, to add and to subtract by
means of playing cards, which were among his first
playthings. Formal arithmetic was begun when he
was 7 years old by spending about an hour a week
upon it. This year he is giving about an hour each
258 INTELLIGENCE OP SCHOOL CHILDREN
week to algebra and about as much to geometry, with
his father as teacher. He has no difficulty with either
subject. Often sets himself problems in geometry to
"In the study of music has applied his ability to
analyze and arrange so that he has made big strides in
musical theory and wants to compose melodies to fit
the words of the poems he selects.
" Conscientious. Refuses to lie. Clings tenaciously
to a standard which he recognizes as desirable. Used
to mark himself for what he considered good writ-
ing and was quick to acknowledge poor work. Obeys
instructions regarding errands, etc. Above average
in unselfishness. Makes plans to give pleasure to
others, and often, with a manifest effort, of his own
volition leaves the best or biggest for some one else.
Loves to share his pleasures. Will remark at selfish-
ness in others."
The above account contains so many things it is
hard to associate with the chronological age of 7 years
that the reader may be inclined to allow something for
maternal prejudice. To do so in this case would be a
mistake. The Binet test, made under extraordinarily
unfavorable conditions, indicates a level of mental
ability not far below that which is normal for children
of 14 years. We have also the testimony of the kinder-
garten and playground instructors in the Horace
Mann school, which agreed thoroughly with the notes
FORTY-ONE SDPEBIOR CHILDEEN 259
furnished by the mother. The average rating given by
the mother on the twenty traits was 1.93; that of the
kindergarten teacher who knew D. best, 1.90. His
former kindergarten teacher says: "D. is a most re-
markable boy. His greatest difficulty has been social
adaptability, but his experience in kindergarten and
playground has brought him well up toward the nor-
mal in this respect. Reads the Iliad and Shakespeare
and publishes a weekly playground newspaper."
One who desired further proof of D.'s exceptional
intelligence would find it in convincing abundance in
any issue of this newspaper, which is a rare essay in
journalism for a boy of 7 years. It is a one sheet,
three column affair, typed. All of the composition is
done by D. who " prints " it on his typewriter. There
is a joke section, an advertising section, a news sec-
tion, and various extras and incidentals from time to
time. The jokes are often such as would not be un-
derstood by children below the mental level of twelve
It will be seen that D. is far superior in general
ability to any of the other children we have described.
His ability seems to compare favorably with that of
Francis Galton, 1 who in childhood showed similar in-
dications of genius. Whether the promise of the pres-
ent will be fulfilled, only the future can tell. How-
1 Terman, Lewis M., "The Intelligence Quotient of Francis
Galton," American Journal of Psychology, 1918. Compare also
"The Psychology of a Prodigious Child," Journal of Applied
Psychology, 1917, by Leta S. Hollingworth and others.
860 INTELLIGENCE OF SCHOOL CHILDREN
ever, considering his fine balance of personal, moral,
and intellectual traits, there is every reason to be-
lieve that he will become a distinguished man.
Indications of superior endowment. Doubtless the
reader has sensed a degree of monotony in the above
descriptions of superior children. Such children show
the usual individual differences in temperament and
personality, but intellectually they have much in com-
mon. Certain qualities are mentioned again and again
by both parent and teacher. Phrases most often used
in giving indications of superior endowment are the
" Alert beyond his years " ;
" Has such keen powers of observation " ;
" Shows a passionate desire to learn " ;
'* Asks endless questions " ;
" Is interested hi everything " ;
*' Is ambitious to excel " ;
" Gets the highest school marks " ;
" Writes such wonderful examination papers ** ;
" Has such a fine command of language " ;
" Has fine reasoning powers " ;
" Shows independence of judgment M ;
" Is an original thinker " ;
"Answers always to the point";
"Has a keen sense of humor" ;
" Has unusual power of concentration " ;
" Is more dependable than other children of his age " ;
FORTY-ONE SUPERIOR CHILDREN 261
" Conscientious to a fault " ;
" Such a lovable child "; etc.
Many are also described as exceptionally truthful,
sympathetic, generous, thoughtful of others, and en-
dowed with a sense of moral responsibility which
shows itself in a willingness to work and to deny them-
selves for social ends. Other symptoms of superior en-
dowment receiving frequent mention include the early
learning of nursery jingles; ease of memorizing; learn-
ing, without instruction, to count and to name the
days of the week and the months of the year; rapidity
of learning to read; learning to read without instruc-
tion by means of newspapers, advertisements, or tele-
phone books; desire to write; love of reading; prefer-
ence for worth-while books; liking for dictionaries and
encyclopaedias; absorption with hobbies, such as col-
lections, wireless telegraphy, and educational games.
These indications are mentioned so often as to appear
well-nigh universal with this class of children.
Only a few have traits that are undesirable. Several
are more or less nervous, a few are exceptionally timid,
three or four are somewhat vain, a few dislike the rou-
tine and restraint of the school, one is rather lazy, one
lacks affection, one shows symptoms of incorrigibility
at home, and several are below average in leadership
and social adaptability. Making proper social adjust-
ments is perhaps the most difficult problem for these
superior children. Their intellectual superiority tends
to set them apart from children of their own age, while
262 INTELLIGENCE OF SCHOOL CHILDBEN
they are at the same time prevented from equal asso-
ciation with older children both by their lack of physi-
cal strength and by the relative immaturity of their
play instincts. No. 42, for example, who at the age
of 7 tests above ISi, obviously cannot compete with
average 13-year-old boys in the usual games of physi-
cal skill, nor is he near enough adolescence to share
their mental outlook. His play interests are in many
respects like those of ordinary children of 7 years; yet
he is largely cut off from natural association with such
children by the fact that he speaks a different lan-
guage. His vocabulary is so " grown-up " that his
playfellows often cannot understand what he is talking
about. Considering such difficulties, the wonder is
that only two or three of our superior children are no-
ticeably queer socially, and that only one borders on
Objections to grading superior children by mental
age. The question may be raised whether the diffi-
culty of social adjustment does not constitute a serious
objection to the plan of grading superior children ac-
cording to mental age, since this would associate them
in class work with children who are several years older.
This danger, however, is largely offset by the oppor-
tunities which the playground offers for making con-
genial acquaintances. The injury done by having
such a child recite with children whom he cannot com-
pete with in play must be very slight compared to the
intellectual and moral injury which is wrought by
FORTY-ONE SUPERIOR CHILDREN 263
keeping him always at tasks which are too easy to
command his best efforts.
One solution would be to have the child of excep-
tional ability remain out of school every second or third
year. This would tend to keep him in class with
children of about his own age, while at the same time
repairing a reasonable amount of effort to keep up in
school work. The plan assumes, however, that the
school authorities will allow such a child to skip the
grade which his fellows take while he is out of school.
If this were not allowed, and often it would not be, the
situation would only be made worse. The plan of
" periodic rests " Las the further objection that by de-
priving the child of the social opportunities which the
school offers it would make his isolation more com-
plete. Besides, there are few homes which could be
expected to fill the child's free year with experiences
of real educational value.
-Opportunity classes for superior children. The re-
sponsibility for~tEelpight education of superior children
belongs with the school. If the opportunities now
offered are not suitable, it is the duty of the school to
provide something better. While some relief is fur-
nished by an elastic system of promotion which will
allow the superior child to skip a half grade occasion-
ally, this should be regarded as a makeshift rather than
a final solution of the problem. The contribution of
the school must be more positive and more educa-
tional. If the needs of superior children are to be
264 INTELLIGENCE OF SCHOOL CHILDREN
met, special classes and special courses will have to
be provided. The advantages of such classes are
(1) They allow children to make rapid progress
without skipping vital parts of the subject-mat-
(2) They allow a broadening and enriching of the
course of study because of the larger accom-
plishments possible to superior minds;
(8) They are a discouragement to vanity because
the level of competition is raised and the
measure of a child's success depends upon his
relative standing in the class;
(4) They insure the mental and moral training
which can come only from sustained effort;
(5) They furnish an atmosphere which is intellec-
tually much more stimulating than that found
in the average class;
(6) Since they bring together children of similar
age and attainments, they go far to solve the
problem of social adjustment.
Wherever " opportunity classes " for bright children
have been tried they have proved an immediate and
surprising success. The children are touched by new
life and inspired with new enthusiasm. That two or
three grades are usually covered in one year is perhaps
a matter of secondary importance compared with the
intellectual awakening and the intensification of effort
which suc^cKsses provoke^ The results have been
FORTY-ONE SUPERIOR CHILDREN 265
so uniformly successful that the special class for gifted
children may be considered to have passed the experi-
mental stage. The following illustration is typical :
In February, 1917, an opportunity class was formed
in Louisville, Kentucky. It consisted of 21 children
selected by means of Binet tests. The intelligence
quotients ranged from 120 to 167, fifteen being above
135. The class covered the work of an entire grade in
a half year. "Besides the accomplishment of this
work, the children learned to use with a considerable
degree of freedom 400 words in conversational Ger-
man. They also composed the words and music of
a spring song and an operetta. The class did this
work happily and with ease. Home study was dis-
couraged, except where it was a matter of great desire,
and then it was limited to twenty minutes. , . . In
character and disposition these children are conceded
by all who know them to be superior. They are not
conceited or puffed up by their selection for the
class." Miss Race, from whom the above is quoted,
states that whatever touches of conceit were present
at the beginning of the class were largely eradicated
before the end of the term. 1
A similar class has been conducted in New York
City by Miss May Irwin, 2 and another in Urbana,
1 Race, Henrietta, "A Study of a Class of Children of Superior
Intelligence," Journal of Educational Psychology, 1918, pp. 91-97.
* See Louise F. Specht: "A Terman Class in School No. 84,
Manhattan," School and Society, March 29, 1919, pp. 393-98.
266 INTELLIGENCE OP SCHOOL CHILDREN
Illinois, under the direction of Professor Whipple. l In
both cases the results agreed in a striking way with
those of the Louisville experiment.
Class sectioning according to mental ability. When
the school system is very small, or when other condi-
tions prevent the formation of a special class for the
children of exceptional ability, their needs may be to a
certain extent provided for by the division of the regu-
lar class into three sections: a slow-moving, a normal,
and a fast-moving group. For example, in a second-
grade class of forty pupils the groups might contain
ten, twenty, and ten pupils respectively. These
could be instructed by the same teacher, but as sepa*
rate classes making different progress and doing work
of somewhat different quality. The work of the three
sections could be so organized that their separate
instruction would be by no means an added burden
to the teacher.
This chapter has been largely devoted to descrip-
tions of children of very exceptional superiority.
Probably not more than one child in a hundred tests
above 135 and not more than one in two hundred
above 140. The children who test between 120 and
135 are several times as numerous, and almost equally
in need of special advantages. It is from this group
that the majority of teachers, doctors, lawyers, min-
isters and other professional men and women come.
1 Whipple, G. M., Classes for Gifted Children (School and Home
Pub. Co., Bloomington, UL, 1919); pp. 151.
FORTY-ONE SUPERIOR CHILDREN 267
Special classes for eight or ten per cent of the pupils
are perhaps not feasible and may not be necessary,
but much can be done by the sectioning of classes in
the manner just indicated and by making the system
of promotion more elastic.
INTELLIGENCE TESTS IN VOCATIONAL AND
EDUCATIONAL GUIDANCE l
Educational and vocational guidance inseparable.
Vocational guidance usually receives attention only on
the eve of the child's departure from school. Thus
restricted, it falls greatly short of its possible value.
If the pupil is to be properly trained for his life-work,
as well as directed to it, his education must at every
step take account of his vocational possibilities. That
is, vocational guidance must be preceded by educa-
Previous chapters have shown how frequently the
school errs in attempting to force children through
courses of study which are beyond their intellectual
capacities, and how futile and discouraging such ef-
forts are. A large proportion of children must leave
school with little direct preparation for life, simply
because they are intellectually incapable of mastering
the contents of a curriculum which the school has set
up as theoretically desirable for alL, It is time that the
school should ask not only what it would like to do,
but what it can do. Facts have been presented which
show that the limits of a child's educability can be
1 See also Hollingworth, H. L., Vocational Psychology (D. Apple-
ton & Co., 1916); pp. 808.
TESTS AND VOCATIONAL GUULJAJNUB
fairly accurately predicted by means of mental tests
given in the first school year. By repeated tests these
limits can be determined accurately enough for all
practical purposes by the end of the child's fifth or
sixth school year. This early, at least, vocational
training and vocational guidance should begin.
The end is not merely that of keeping the child in
school. This in itself is not necessarily desirable. In
the conservative school system offering only the tradi-
tional courses, it is perhaps just as well that pupils of
80 1 Q or lower (i.e., 10 per cent of all) should drop out
by the age of fifteen years. By that time they have
gotten about all they can get from the older type of
restricted elementary curriculum. Continuation would
mean nothing more than to remain hopelessly stranded
in the sixth or seventh grade, without further effective
training except training in failure.
Limitations of vocational guidance. It must not be
supposed that vocational guidance, in the sense of de-
termining exactly which of a thousand or more voca-
tions a given individual should enter, is yet possible.
The most important contribution which psychological
tests are at present prepared to make is in the meas-
urement of general intelligence. The special abilities
which so largely influence success in the majority of
vocations have not yet been satisfactorily analyzed,
much less measured. The intangible factors of inter-
est, will power, social adaptability, leadership, and
personality are still less subject to exact determina-
270 INTELLIGENCE OF SCHOOL CHILDREN
tion, although their combined influence upon voca-
tional success is doubtless very great. One's general
ability may fit him equally for success in a dozen
different vocations, and in this case the ultimate
choice should depend upon practical considerations,
natural interests, and various traits of personality.
Nevertheless, intelligence tests will be of great
value in vocational guidance, even if they tell us noth-
ing more than that reasonable success in a given voca-
tion is or is not compatible with the general mental
.ability which an individual possesses. The saddest, as
well as perhaps the most common failures in life are
due to the selection of a vocation which requires a
higher grade of ability than the individual possesses.
Hardly less unfortunate is the person whose too mod-
est self -estimate lands him in an occupation that is in-
tellectually beneath him. A mistake in either direction
entails bitter disappointment, since often it is not dis-
covered until the time for new choices has gone by.
Mistakes of this kind can be largely prevented by
intelligence tests as soon as the proper factual basis
has been laid^JBIrst, however, it will be necessary to
find.the^actual ranges of intelligence represented in
the different types of vocations, and especially the
lower limit of intelligence which permits reasonable
success^ It will also be necessary to determine for each
.typical vocation the level of mental ability which rep-
resents the " point of diminishing returns," in order to
prevent superior ability from being wasted upon voca-
TESTS AND VOCATIONAL GUIDANCE 271
tions which make only moderate intellectual demands.
Tests in sufficient number will doubtless show that
there exists for most vocations a middle range of men-
tal ability in which the chances of success are near the
maximum, that intelligence below this range becomes
less and less favorable to success until a " dead line " is
reached, and that ability of a higher order represents
only so much sheer waste. When such standards of oc-
cupational intelligence are available, they will furnish
the most important single basis for vocational and edu-
cational guidance. Knowing the intelligence of the
child we could then select the vocations well within
the range of this intelligence, and leave it to the child's
natural interests and to practical considerations to
make the final choice. Such a method would not elimi-
nate the possibility of vocational failure, but it would
eliminate one of its most common causes. Until the
intellectual requirements of ..the different vocations
have been more definitely established, some sugges-
tion for guidance may be gleaned from the following
studies of typical vocational groups.
Firemen and policemen. In 1916 the city of San
Jos6, California, made an unusual experiment, per-
haps the first of its kind in this or any country. 1 The
experiment involved a civil service examination for
positions in the fire and police departments, based
entirely upon standardized mental and educational
1 Tennan, Lewis M. t "A Trial of Mental and Pedagogical Tests
in a Civil Service Examination for Policemen and Firemen,"
Journal of Applied Psychology, 1917, pp. 17-29.
272 INTELLIGENCE OF SCHOOL CfflLDBEN
tests. The tests used included the Stanford-Binet in-
telligence scale, the Trabue Completion Test, the
Thorndike Oral Reading Test, the Courtis Standard
Test in arithmetic, a handwriting test, and the Otis
tests of spelling and arithmetical reasoning.
Thirty candidates presented themselves in competi-
tion for the ten or twelve prospective openings. All
were American born, with ages distributed fairly evenly
between 21 and 38 years. Their incomes during the pre-
vious year ranged from $420 to $1350, with a median
of $960. Their previous occupations ranged from
totally unskilled to skilled and high-grade clerical.
The distribution of mental ages and I Q's was as
The median mental age was 13-5, the median I Q 84. *
The lowest fourth fell below 78 1 Q, the highest fourth
reached 91 1 Q or above. The minimum I Q compati-
ble with efficiency for policemen and firemen is not
known, but, in the absence of a definitely established
standard, all who tested below 80 I Q were rejected
1 The reader is reminded that in calculating the I Q for adult
subjects chronological age above 16 years is disregarded. That is,
mental age is always divided by 16.
TESTS AND VOCATIONAL GUIDANCE 273
without further consideration. Choice from the re-
mainder was made on the basis of personal history,
and on the combined results of the various tests.
Among those testing below 80 1 Q were four individ-
uals who were already serving in the fire department
as " extras." They had gotten their positions under an
earlier political regime. The I Q's of these four men
were 63, 74, 77, and 79. The 63 I Q individual was 34
years of age, and had never earned a wage more than
two thirds as high as that paid the average unskilled la-
borer in his community. His deficiency is well known
to his acquaintances, and he had secured his position
as " extra M only through the influence of his father, a
man of some local prominence.
The individual who tested at 67 1 Q was pronounced
by the captain of his militia company to be unques-
tionably feeble-minded. He had never done better
than unskilled labor, and at the time of the examina-
tion was without employment. Another of 71 1 Q had
formerly worked as a hotel porter and also as a railroad
signalman. Although the duties of a railroad signal-
man are extremely simple, they require attentive per-
formance, and one may well doubt whether they can
be safely entrusted to an I Q of 71.
The following notes may also be of interest as show-
ing what may be expected of various I Q's:
I Q 77. Common laborer in a sawmill. Had served
one term in the regular army, and reenlisted after
failure in the examination.
374 INTELLIGENCE OF SCHOOL CHILDREN
I Q 78. Deliveryman for a grocery store and "ex-
tra " in the fire department.
I Q 78. A teamster, unskilled laborer.
I Q 79. No occupation except as " extra " in the
I Q 81. Had served several years as policeman in
an eastern State at $65 to $80 per month.
I Q 83. A successful street-car conductor, said to
be very popular with his patrons because of his genial
good nature and his interest in people.
I Q 112. Had completed the second year of high
school, and had earned as high as $135 per month as a
salesman. His purpose in securing a position in the
fire department was to secure leisure for a correspond-
ence course in expert accounting.
Such data would suggest that the I Q of 75 or below
belongs ordinarily in the unskilled labor class, that 75
to 85 is preeminently the range for semi-skilled labor,
and that 80 or 85 is ample for success in some kinds of
skilled labor. When the candidates were classified into
unskilled, semi-skilled, and skilled, according to the
occupations they had followed, the following I Q
ranges and averages were found:
Skilled or better
Range of I Q
63 to 89
74 to 96
84 to 112
Average I Q
TESTS AND VOCATIONAL GUIDANCE 275
Express company employees. Flanders 1 gave Stan*
ford-Binet tests to 47 employees of a large express
company. Only those were tested who had been with
the company at least a year. The work they were do-
ing is indicated by the following random selections:
Accounting Clerk, C.O.D. Clerk, Settlement Clerk,
Waybill Clerk, Receiving Clerk, Clerk in Value Room,
Clerk in On Hand Department, Wagon Dispatcher,
Chief Router, etc. These were typical of the rank and
file of seven hundred employees, not including super-
visors or semi-officials at one end or floaters at the
other. In practically all cases the work involved a high
degree of specialization, "offering exceedingly limited
opportunity for the exercise of ingenuity or even per-
sonal judgment. Success is achieved by the faithful
and careful performance of a simple task for the doing
of which perfectly definite rules have been given." Al-
though the work done by the forty-seven employees
apparently differed little as regards the amount of in-
telligence required, the following wide range of mental
ages was found:
The range was from 10 years (I Q 62) to 18-7 (I Q
116) with a median of 15-2 (I Q 95). One fourth were
1 Flanders, J. K., "Mental Tests of a Group of Employed Men,
Showing Correlations with Estimates furnished by Employer,"
Journal of Applied Psychology, 1918, pp. 197-206.
376 INTELLIGENCE OF SCHOOL CHILDREN
below 13-10 (I Q 86) and one fourth above 16-7 (I Q
It is surprising to find men with intelligence which
would enable them to take a college course competing
with others who could never graduate from the eighth
grade. As stated by Flanders, " such individuals are
possibly lacking in certain emotional, moral, or other
desirable qualities; it may be that economic pres-
sure crowded them out of school before they were
able to prepare for more exacting service; it may be
that the schools did not provide them with suitable
vocational training; it may be that they selected their
vocations blindly and ignorantly . "Whatever the reason
there is evidently a big social and economic loss."
Flanders concludes by calling attention to the abun-
dant occupational opportunities open to men of 70 to
80 I Q (mental age 11 to 13 years). The evolution
of modern industrial organization together with the
mechanization of processes by machinery is making
possible a larger and larger utilization of inferior men-
tality. One man with ability to think and plan guides
the labor of ten or twenty laborers, who do what they
are told to do and have little need for resourcefulness
or initiative. It is even suggested that our chief diffi-
culty may soon be to provide enough suitable jobs for
those of higher intellectual capacity. We can at least
rest assured that society has and will continue to have
place enough for workers of decidedly inferior intelli-
gence provided they are given a training which is suf-
ficiently practical and concrete.
TESTS AND VOCATIONAL GUIDANCE ' 277
Street-car employees and salesgirls. C. W. Waugh
tested 82 street-car motormen and conductors, 61
salesgirls in a large department store, 7 railroad engi-
neers, and 4 department store " buyers." The men-
tal ages found for these groups were as follows:
The medians were as follows:
Street-carmen 13-8 (IQ85.6)
Salesgirls 1S-6 (IQ84.5)
Engineers 16.0 (TQlOO)
Buyers 17.0 (IQ106)
The work of a street-car motorman or conductor
rates as semi-skilled. The investigation showed that
an I Q of 80 to 90 is entirely satisfactory for this kind
of work provided other traits are favorable. However,
a study of the ratings given the men for efficiency indi-
cated that a 75 I Q is an unsafe risk either for motor-
man or conductor. The one testing lowest, 10-5, 1 Q
65, had a low efficiency rating l and at the time of the
test was laid off because of a serious accident caused
1 Unfortunately the original data of this investigation were lost
before the correlations of I Q with ratings could be computed.
278 INTELLIGENCE OF SCHOOL CHILDREN
by his carelessness. On the oth3r hand, the data
suggested that intelligence above 00 or 100 I Q adds
nothing to the efficiency of a motorman or conductor,
and that it conduces to discontent. Most of those of
highest I Q stated that they were only engaged in the
work because of bad luck or unfavorable labor condi-
tions and that they looked forward to getting some-
thing better. Men testing around 80 or 85 usually
seemed contented and proud of their jobs.
The work done by the salesgirls would rate all the
way from unskilled to semi-skilled, or in general
slightly lower than the work of street-car conductors
and motormen. The I Q distribution for salesgirls,
however, was about the same as that for street-car
men. This is another illustration of what is probably
generally true in our present industrial organization,
that the economic situation for men of a given I Q is
considerably easier than for women of the same intel-
lectual ability. The data for motormen, conductors,
firemen, and policemen indicate that an I Q of 85
among men receives about the same economic rewards
as an I Q of 100 to 120 among women, taking the
average elementary teacher, or high-grade stenogra-
pher as typical of this class.
Business men. Knollin and Zeidler tested 30 busi-
ness men of moderate success and limited educational
advantages. The subjects were typical of the kind of
men who own or manage the ordinary stores, barber
shops, draying business, etc., in a small town. None
TESTS AND VOCATIONAL GUIDANCE 279
had graduated from a high school, and only two had
attended school above the eighth grade. None had
accumulated any considerable fortune, and none had
failed outright in business. The following mental
ages were found:
13 to 13-11
14 to 14-11
15 to 15-11
16 to 16-11
17 to 17-11
18 to 18-11
The median mental age was 16-2 (I Q 102). The
lowest fourth were below 15-0 (I Q 93.6), and the
highest fourth above 17.2 (I Q 107). The only indi-
vidual testing below 14 runs a successful delicatessen
establishment. There is no doubt about his inferior
intelligence (I Q 81), but he is exceptionally industri-
ous and is aided by a wife who is reputed to be " the
brains " of the business. This was the only I Q below
88. The group as a whole presents an interesting con-
trast with the unskilled and semi-skilled groups tested
by Waugh and Flanders.
Tests of college students. Stanford-Bine! tests
were given under the direction of Coover to 62 stu-
dents in a psychology class at Stanford University.
The group was fairly representative of the student
body above the freshman year. The distribution of
I Q's was as follows:
280 INTELLIGENCE OF SCHOOL CHILDREN
The median I Q was 113. One fourth tested below
108 and one fourth above 117. The lowest I Q was 94.
Dr. June Downey tested 42 freshmen and 49 upper-
classmen of the University of Wyoming. 1 The me-
dian scores for the three groups were: Freshmen,
16-8 (I Q 104); upperclassmen, 17-2 (I Q 108). The
I Q distribution for all of Dr. Downey's students taken
together was as follows:
Dr. Downey found for the members of her psychol-
ogy class a correlation of .527 between I Q and her
own estimates of intelligence (previously made). In
regard to the relative accuracy of the tests and her
ratings, Dr. Downey adds: " More intimate acquaint-
ance with the class convinced me, moreover, that the
I Q's were much more accurate than my unaided judg-
ment. In a number of instances I was able to deter-
mine just the factor that had led me astray."
In a majority of cases the results of the tests agreed
fairly well with class marks. Of the seven freshmen
who tested below I Q 94, only one returned for work
the following year. "This is her third year in the
freshman class, a hopeless drifting from one depart-
ment to another."
1 Downey, June E., "The Stanford Adult Intelligence Tests,"
Journal of Delinquency, 1917, pp. 144-55.
TESTS AND VOCATIONAL GUIDANCE 281
As would be expected, some students did much bet-
ter or much poorer work than the I Q would suggest.
The following are typical cases of such disagreement:
" Young man. Passed every test in the scale, but
is noted for his many failures in courses. His reputa-
tion in college is that of a young man of ability who
chooses to turn his talents in other than academic
"A girl, whose very poor work led us to expect a
record very much lower than she gave. Shyness and
indifference are, I believe, the cause of her poor work.
A little extra attention in class convinced me of the
accuracy of the test results." She is described as
having ability to give "precise and brief answers to
questions " and to " hit the nail on the head once her
interest is aroused."
The tests of college students justify the conclusion
that the student bodies of colleges and universities are
recruited mainly from those whose intelligence is con-
siderably above the median for people in general. This
is true to an even greater extent than the I Q's found
would indicate, since, as we have explained elsewhere
(p. 147) the Stanford-Binet does not adequately meas-
ure adults of exceptionally superior ability. In all
probability the large majority of college students
would as children in the grammar grades have tested
between 100 and 130, with a median of perhaps 115
to 120. A certain number would probably have
tested between 90 and 100, but the chances are remote
882 INTELLIGENCE OF SCHOOL CHILDREN
that a child testing much below 90 will ever be able to
satisfy the requirements for college graduation. Chil-
dren who test below 100 should ordinarily not be en-
couraged to look forward to entrance into law, medi-
cine, the ministry, engineering, teaching, or any other
profession which demands a high degree of ability in
abstract or conceptual thinking. Substantial success
in such professions is probably achieved only by indi-
viduals above the 115 or 120 I Q class.
Tests of social and industrial failures. Knollin
tested 154 " migrating unemployed " men who sought
temporary shelter at the " hobo hotel *' of Palo Alto,
California. Many of these were tramps by profession;
some were merely traveling by foot to other parts of
the State in search of employment. The mental ages
found were as follows:
The median mental age was 14-2 (I Q 89). The
lowest 5 per cent were below 12-7 (I Q 79), the high-
est 25 per cent above 15-8 (I Q 98).
Johnson gave the Stanford-Binet tests to 107 desti-
tute men picked at random from the unemployed cared
for by various social service organizations in Portland,
Oregon. 1 The following mental ages were found:
1 Johnson, Glenn R., "Unemployed and Feeble-Mindedness,"
Journal of Delinquency, 1917, pp. 59-73.
TESTS AND VOCATIONAL GUIDANCE 283
It will be seen that the proportion of low-grade cases
is larger in the unemployed groups than for conductors,
motormen, salesgirls, or express company employees.
About 5.5 per cent of Johnson's group test below 10
years, and 12 per cent below 11 years. The corre-
sponding figure for Knollin's group are 5.2 per cent
and 8.4 per cent. However, so many of the unem-
ployed have average or superior intelligence that the
median mental age for the two unemployed groups
combined is 14-3, and the median I Q 89. This
median exceeds that found for street-car men, sales-
girls and the San Jos6 civil service applicants, but is
considerably lower than the median for business men
and railroad engineers.
From the point of view of vocational education and
vocational guidance the above facts are very signifi-
cant. Plainly, unemployment in the case of the large
majority of these men is not accounted for by their
lack of intelligence. More than 60 per cent had intel-
ligence fully equal to that of the average of the 82
regularly employed street-car employees. At least
10 per cent of them were the intellectual equals of the
average Stanford University student, and probably
25 per cent were intellectually capable of graduating
from a high school.
Even prisoners and juvenile delinquents, among
284 INTELLIGENCE OF SCHOOL CHILDREN
whom the proportion of feeble-mindedness is admit-
tedly high, are more often than not well within the
bounds of intellectual normality. Prom the scores of
studies of prison and reform-school inmates, the data
of Williams may be presented as typical. The men-
tal ages found among 184 delinquent youths over six-
teen years of age at the Whittier State School were as
The median mental ages for these delinquents is
12-6, and the median I Q 78. Probably two thirds
of the entire number are intelligent enough to make
good unskilled workers. Similar facts were found in
the case of 150 consecutive entrants at the San
Quentin State Prison, California, who were tested by
Knollin. Nearly half of the prisoners were equal in
intelligence to the average street-car employee (semi-
skilled labor) while several were as intelligent as the
average college student.
Those who have made psychological studies of juve-
nile delinquents, prisoners, and the unemployed have
placed the emphasis upon the large amount of feeble-
mindedness found. All will admit that a large propor-
tion of both groups are defective or border-line cases,
perhaps 20 or 25 per cent of prison and reform-
school inmates and possibly 10 per cent of those out of
TESTS AND VOCATIONAL GUIDANCE 285
employment .in an average city under average eco-
nomic conditions. It would be a serious mistake,
however, if our concern over the necessity of social
control for defectives should lead us to overlook the
large majority in both groups who, as far as intelli-
gence is concerned, may be considered potential social
assets of great value. It would be interesting to know
to what extent the failure of such individuals could be
prevented by such measures as vocational education,
vocational guidance, and courses of study sufficiently
differentiated to fit the abilities and to satisfy the
interests of all the children who are above the dead-
line of mental deficiency. It will be noted that 45
per cent of Johnson's unemployed and not far
from 70 per cent of the delinquents fall within the
range 70 to 89 I Q. This is the range which fur-
nishes the majority of school dullards. When we
investigate the school histories of men who test be-
tween 70 and 80 we are almost certain to find a record
of low marks, failure, and serious retardation. Those of
the 80 to 90 class have usually failed less seriously, but
have rarely shown the ability to get much beyond the
eighth grade. The majority of the 70 to 85 class have
left school between the fifth and eighth grade with little
preparation for life or life's work. It is no wonder that
many fail and drift easily into the ranks of the anti-
social or join the army of Bolshevik discontents.
For convenience the I Q distributions of the various
vocational groups described in this chapter are brought
86 INTELLIGENCE OF SCHOOL CHILDREN
together in Table 40. The scores of 6 railroad engi-
neers and 4 department-store " buyers " are thrown in
with those of Knollin's 30 business men. The data
for Knollin's hoboes and Johnson's destitute men are
also combined, as the I Q distributions were about the
same for the two groups. For each group the median
I Q is given, also the I Q which marks off the lowest
fourth of those in the group.
Per cent in each I Q group
104 I Q
<Y7 T Q
ft JL \l
ductors. . . .
79 I Q
78 I Q
77 I Q
71 I Q
TABLE 40. I Q DISTRIBUTION OF VARIOUS VOCATIONAL GROUPS
Educational guidance. In vocational guidance the
best that intelligence tests can do is to indicate roughly
the vocational level in which success is possible. The
final choice of a vocation must be determined largely
by interest and opportunity. For all we know, law,
medicine, engineering, teaching and the ministry make
about equal demands upon general intelligence. Per-
TESTS AND VOCATIONAL GUIDANCE 387
haps carpentry, masonry, plumbing, blacksmithing,
etc., require about the same amount of intelligence as
dozens of other skilled trades. However, intelligence
tests can tell us whether a child's native ability cor-
responds approximately to the median found in the
professions, the semi-professional pursuits, the ordi-
nary skilled trades, the semi-skilled trades, among un-
skilled laborers, etc., and this information is of great
value in planning a child's education. Itjs^aqcordr,
ingly in educational guidance that intelligence tests
EaveTtHeiip chief value.
Tables 28 to 35 (pp. 159-162) have shown the school
progress that may be expected of various grades of
intelligence, and the facts set forth in other chapters
have indicated the relation of intelligence to elimina-
tion and to the ability to master high-school or college
courses. The universal testing of school children
would save many a disappointment, A certain woman
of intelligence and education has a daughter who at
the age of seventeen years tested at 78, and was still
in the seventh grade. Yet the mother had not given
up the hope that the daughter might become a sten-
ographer. A college professor with a twelve-year-old
son who tested at 83 was planning to send him through
college. The boy will be fortunate to complete the
eighth grade. Such children are sometimes badgered
and urged on until life is a burden. The son of a cer-
tain lawyer has always tested at 80 to 85. He wishes
to become a gardener, and his profitable success in
288 INTELLIGENCE OF SCHOOL CHILDREN
tilling numerous vacant lots sufficiently attests his
ability in this line. The father, however, insists that
his son must have a college education. To this end
he scolds, coaxes, and employs private tutors. His
best efforts, however, have only brought the boy to
the second year of high school at the age of twenty
years. The boy comprehends nothing that he is
taught and keenly dislikes school.
On the other hand, it is by no means uncommon for
exceptionally bright children to be apprenticed early
to occupations which require but mediocre intelligence.
Anything above 85 1 Q in the case of a barber probably
represents so much dead waste; yet we know a barber
who is as intelligent as the average college student.
Although in our country the industrial lines of cleav-
age are not rigid enough to prevent ready shift from
one occupation to another, provided one determines
to make the shift, it must not be forgotten that, after
all, men are largely creatures of habit and after a cer-
tain age do not find it easy to adjust to the require-
ments of a new vocation. If we knew the total waste
of mental ability we should probably be appalled. The
waste is probably enormous in the case of women,
because of the limited number of vocational oppor-
tunities open to them.
^ The conservation of talent. A nation's intellectual
assets are the most precious it will ever have, and the
principle of conservation will find here its most useful
TESTS AND VOCATIONAL GUIDANCE 289
In the conservation of talent the teacher occupies a
strategic position. It is her duty to foster in a pupil
the highest ambitions which are consonant with his
intellectual endowment. To expect that she will be
able to estimate a pupil's endowment accurately
enough by mere observation is to expect too much.
We have known so many bright children who were
seriously underrated by their teachers that the neces-
sity of the test method, as a supplement to observa-
tion, seems hardly open to question. If tests were
more commonly given we should probably find many
children like the following:
A. B. was twelve years old and in the sixth grade.
He was failing, or at any rate his work was unsatis-
factory to the teacher. As a matter of fact she did
not promote him at the end of the term. The father
consulted Mr. Virgil E. Dickson, the psychologist of
the city schools, who gave the boy a mental test. The
I Q was approximately 140. Apparently A. B. had a
grade of ability not equaled by more than one child in
two hundred. Inquiry disclosed the fact that the boy
had formed a dislike for his teacher. This teacher
required her pupils to copy from the dictionary the
definition of all the new words encountered in each
lesson. When A. B. said he knew all the words she
accused him of untruthfulness. In reality his vocabu-
lary was equal to that of the average teacher. The
case deserved radical treatment and got it. Mr. Dick-
son, notwithstanding the boy's non-promotion in the
290 INTELLIGENCE OF SCHOOL CHILDREN
sixth grade, arranged for him to skip both the seventh
and the eighth grade and to enter high school immedi-
ately! He did so, and passed all his work with good
marks. For some months his teachers were not told
of the heresy that had been committed, and they never
suspected that the pupil had not come to them in the
usual way. Cases of this kind suggest an explanation
for the traditional but incorrect belief that a majority
of great men and women were dull or mediocre in
PRACTICAL SUGGESTIONS FOR THE USE OF MENTAL
Teachers must learn to use tests. Unless the rank
and file of teachers learn to use tests the universal
grading of children according to mental ability -will
remain largely a Utopian dream. We cannot agree
with those who hold that Binet tests should not be
attempted by teachers. Teachers are universally
encouraged to use such educational measurements as
the Courtis tests, handwriting scales, and tests of
ability in reading, history, and composition. Yet, it
is fully as difficult to learn the correct procedure for a
" battery " of six or eight standard educational tests
as to acquire a reasonable facility in the use of the
Binet scale. It should be emphasized, however, that
wherever possible the use of both educational and in-
telligence tests should be supervised, either by a psy-
chologist or by some one else who has had extended
experience in their use and in the interpretation of
results. It is here that the psychologist finds his
proper task, rather than in giving the tests himself.
The public school psychologist, for example, cannot
himself give more than 800 to 1000 Binet tests in a
1 See also Terrain, Lewis M., The Measurement of Intelligence
(Houghton Mifflin Co., 1916); pp. 362.
892 INTELLIGENCE OP SCHOOL CHILDREN
school year, but he can supervise the testing and
grading of 10,000 by taking advantage of the help
which teachers can give. This method not only in-
creases the number of pupils who will be graded more
nearly in accordance witl* their abilities; it also effec-
tively stimulates the teacher's interest in her children.
Preparation needed for Binet testing. However,
no one should attempt to use the Binet scale without
careful preparation. The training needed can be
given effectively in the normal school. A half-year
course of three lessons per week, or a somewhat shorter
course with five lessons per week, will serve the pur-
pose if it is properly supplemented by other courses
in educational psychology. Such a course should (1)
introduce the student to the nature and extent of
individual differences among children; (2) show the
bearing of these upon school grading; (3) explain the
fundamental principles underlying intelligence test-
ing; and (4) give a fair degree of mastery of the Binet
procedure. The use of at least one scale for group test-
ing should als6 be taught. The course should include
actual testing by the student as well as demonstration
tests. Courses of this kind should be considered an
indispensable part of the normal-school curriculum.
Meanwhile, what about the teacher in service who
has not had the advantage of such instruction? Must
she continue to rely on guesswork for the classifica-
tion of her pupils? A fairly extended experience has
convinced us that this is not necessary. With a little
THE USE OF MENTAL TESTS 293
help the average teacher can prepare herself to use
standardized mental tests accurately enough for prac-
tical purposes. If there is a psychologist in the school
system the problem can be solved by forming Saturday
or afternoon classes for givirtg the needed instruction.
If no psychologist is available, some one else can often
be found in the school system who is capable of
directing the work, perhaps the director of special
classes, or a principal who has had some training in
the use of tests. A six-weeks course in a summer ses-
sion of a normal school or university will also provide
the necessary training.
How to learn the Binet procedure without instruc-
tion. If no help is available the earnest teacher need
not hesitate to undertake the task alone. It is best to
begin by first mastering the contents of two or three
books dealing with individual differences and the
principles of mental testing. Then the Binet pro-
cedure should be carefully studied. Merely to read
through a description of the tests is not sufficient.
The directions should be studied with the closest
attention to the finer details of procedure, and to the
method of scoring, computing mental age, etc. The
significance of mental age as a basis of school grading
and of the I Q for forecasting a child's later develop-
ment should receive special attention. Actual testing
may then be begun, preferably with the help of the
Record Booklet prepared for use with the Stanford-
Binet Scale. This contains appropriate spaces for
294 INTELLIGENCE OF SCHOOL CHILDREN
the verbatim recording of results, and gives abbrevi-
ated directions for scoring. It also saves memory
strain and prevents error by supplying the exact word-
ing for many of the tests.
After testing one or two children the instructions
should be read through again, and each step in the
procedure noted. This always brings certain points
into relief which have escaped notice. By thus
checking up her procedure after each test the teacher
will acquire a sureness and a degree of accuracy which
nothing else can give. After fifteen or twenty tests
the mental age secured should be substantially the
same as a trained psychologist would get, assuming
that the teacher has studied the directions with con-
scientious care and has learned not to take liberties
It is necessary to understand, from the very begin-
ning, that an apparently trivial alteration of a test
may so change its nature as to invalidate the results.
The formula for each test should be adhered to strictly.
Questions should ordinarily not be repeated except
when the instructions indicate that it is permissible
to do so. It is necessary to avoid leading questions
and other fonnsf of unintentional aid. If the child's
answer is not dear, the question " What do you mean? '*
is usually sufficient and is practically the only form of
supplementary question allowable.
A free and easy manner with the child should be
assiduously cultivated. Timidity must be overcome
THE USE OF MENTAL TESTS 295
so that the child will do his best without undue coax-
ing. If the child's confidence is to be gained, it is
necessary to take his efforts seriously, however absurd
they may appear. Attention should never be called
to errors. Apart from such some vague commenda-
tion as " You have done beautifully," etc., the child
should be told nothing of the result of the test.
The examination should be thorough. It should
extend down the scale far enough to include at least
one year in which there is no failure, and up far enough
to include at least a year in which there is no success.
By using only the six regular tests in each year the
examination can ordinarily be completed in thirty to
forty minutes with younger subjects, or in fifty minutes
with older ones. With subjects of the high-school
level a little more time is occasionally necessary. At
first the time is prolonged by the recording of replies,
which should always be done as nearly as possible
verbatim. A little experience and the liberal use of
abbreviations soon enables one to do this without
retarding the examination appreciably.
Learning to score. The responses, if recorded,
should be scored immediately after the examination
has been completed. If responses are not recorded,
the scoring must be done as the examination proceeds.
Wherever there is the slightest doubt as to the satis-
f actoriness of a response, the guide should be consulted
and followed. Since the scale has been standardized
on a definite basis of scoring, it is evident that unless
296 INTELLIGENCE OF SCHOOL CHILDREN
this rule is adhered to, the resulting mental age and
I Q lose their significance. The teacher must learn
to suppress her personal judgment as to how a test
ought to be given or scored, and to ask only the ques-
tion how it is given. With conscientious effort the
errors of scoring can soon be reduced to a reasonable
If a psychologist is available, the teacher's scoring
should be checked up from the written responses until
the right habits have been thoroughly established.
This is the method followed by Dickson, who writes
as follows regarding the errors made by twenty-one
Oakland teachers in scoring several hundred tests:
** Before the testing was begun, six lessons of 1 J hours each
were given. Each teacher then tested her own pupils, and
graded and marked her own tests. I then graded the tests
myself, with the following results:
No correction of mental age necessary 68 . & per cent
Correction of months ............. 20 . 4 per cent
Correction of 4 months .............. 10.8 per cent
Correction of 6 months .............. 0.6 per cent"
This excellent record is explained in part by the fact
that the subjects were all first-grade pupils, so that the
teachers were not compelled to learn the procedure
and scoring for the tests below four years or above
ten. It will be noted that hardly any errors necessi-
tated a correction of more than 4 months in mental
age. For the average first-grade child an error of this
THE USE OP MENTAL TESTS 29?
amount would affect the I Q to the extent of only 5
We have elsewhere reported the errors of five univer-
sity students in scoring 843 Binet tests. 1 The mental
ages as computed by the students were correct
within 2 months in 84.8 per cent of cases;
within 4 months in 95.5 per cent of cases;
within 6 months in 98.6 per cent of cases.
The average error in I Q was about 1 point. Approxi-
mately one third of the necessary corrections were due
to arithmetical mistakes in counting the number of
"plus*' marks, adding months of credit, or dividing
mental age by chronological age. Practically all of
the errors of more than G months in mental age, or of
more than 5 to 8 points in I Q, were of this preventable
kind. The counting and adding of credits and the divi-
sion for I Q should always be done twice for each sub-
Tabulation of the errors in scoring the separate
tests in the scale showed that two tests were responsi-
ble for 30 per cent of the errors. These were the ball
and field test and the description and interpretation
of pictures. Others which gave rise to frequent errors
were the following: definitions by use and superior
to use, interpretation of fables, the comprehension
questions, the diamond, designs, and definitions of ab-
1 Terman, Lewis M., "Errors in Scoring Bitict Tests/' The
P*y<:lit>l<>i/iMl Clinic, UU8, pp. &
298 INTELLIGENCE OF SCHOOL CHILDREN
stract terms. The directions set forth in The Meas-
urement of Intelligence for scoring these tests should
be consulted again and again until they have been
The Interpretation and use of results. To acquire a
reasonable degree of expertness in giving Binet tests is
a much simpler matter than to learn how to interpret
and use the results. We have written this book prima-
rily to show concretely the significance of mental age
and intelligence quotient in the grading of school chil-
dren. While its careful study should aid the amateur
to avoid gross errors in the use of results, there is
much which experience alone can bring and much
which only those of psychological training can ac-
quire. In cities which employ a school psychologist
the problem is simple enough; the teachers can make
the tests and leave it to the psychologist to interpret
the results and to utilize them in the classification
of children. If there is no recognized expert in the
school system the teacher must work with caution.
She must learn to consider her interpretation of the
test in a tentative light, and must avoid the risk of
passing judgment in doubtful or apparently patho-
logical cases. She must understand clearly that the
mere ability to give a Binet test acceptably gives her
no claim to the title " clinical psychologist/' If she will
use the test simply as a means of getting a more ac-
curate idea of a child's mental ability than she could
get in any other way, she will be amply rewarded.
THE USE OF MENTAL TESTS 299
For obvious reasons the teacher should use discretion
in talking about the results of the tests. That the
child should not be told his mental age or I Q has al-
ready been emphasized. The teacher will also find
that it is generally unwise to discuss the test results
with parents in very specific terms. Such expressions
as " exceptionally bright," " mentally retarded," or
"slow to learn," are usually harmless; but expressions
like "dullard," "feeble-minded," "border-line," etc.,
should be avoided. Even if the parents know the child
to be feeble-minded they resent the teacher's saying
so, justly feeling that the diagnosis of mental defect
is not within her province. This is the rule, but of
course there are exceptions. The tactful teacher who
has the confidence of the mother can sometimes talk
with her quite frankly about the defects of her chil-
dren. The teacher's attitude should always be one of
sympathetic helpfulness. Levity or cynical remarks
about the dullness of a pupil should always be avoided,
It is best not to discuss I Q's and mental ages of in-
dividual pupils too freely among acquaintances or
even among colleagues. One never knows when or
where a chance remark will be repeated.
Above all, the teacher must learn not to interpret
the results of her tests too literally and not to depend
upon them too exclusively. The child is not all intel-
ligence; his fitness to take up the work of a given grade
is determined partly by such factors as health, indus-
try, attitude toward school work, and regularity of at-
800 INTELLIGENCE OF SCHOOL CHILDREN
tendance. Immediate and wholesale re-grading of the
school on the basis of mental age as soon as the tests
have been completed is not recommended. It is best
to begin with individual children who are most seriously
misplaced, especially the very bright, who are nearly
always one or two grades below where they belong.
As one after another of these is found to continue to
do good work after extra promotions, the teacher will
gradually acquire confidence in her judgment and in
the verdict of the tests.
It is necessary, however, to avoid the danger of mak-
ing a fetich of the I Q, which we have shown to be by
no means infallible. An I Q of 85, for example, means
no more nor less than that the child tested later will
probably be found between 80 and 90. It does not
mean that he may not later test as high as 100 or as
low as 70, although the chances are roughly 2$ to 1
against his doing so. Because of the possibility of such
errors, however, it is necessary to check up the results
of the tests in every possible way. The test should
mark the beginning, not the end, of the teacher's
study of a given child. As a point of departure the
intelligence test is of great value; accepted as a final
verdict it may lead to mistakes and disappointment.
Children who cannot do the school work within at least
one year of that corresponding to mental age should be
studied. Usually a reason will be found. Perhaps the
child lacks self-confidence. Possibly because of timid-
ity his school work has not shown up at its full value.
THE USE OP MENTAL TESTS 801
Perhaps there has been lack of application. What-
ever explanation is found, the teacher will understand
the child in a way that would never have been possible
without the insight which the test gives. Cases which
continue doubtful or puzzling should be re-tested.
The use of supplementary data. Before beginning
her tests the teacher should secure the following data
for each child:
1. Age, in years and months;
2. Years in school;
8. Record of illnesses;
4. Nationality of each parent;
5. Occupation which supports the family;
6. Data regarding the child's brothers and sisters.
It also greatly enhances the value of test results if
these can be compared with ratings based on observa-
tion. For this purpose the teacher should rate each of
her pupils for quality of school work, general intelli-
gence, and two or three personal traits like dependa-
bility, social adaptability, conscientiousness, etc. The
ratings should be made on the basis of either a five-
fold or seven-fold classification, as follows:
Five-fold classification Seven-fold classification
1. Very superior 1. Very superior
$. Superior & Superior
8. Average 8. High average
4. Inferior 4. Average
5. Very inferior 5. Low average
7. Very inferior
802 INTELLIGENCE OF SCHOOL CHILDREN
The ratings should, of course, be made in advance
of the tests, in order that they may represent an inde-
pendent judgment. Their comparison later with the
test results will prove of surpassing interest. One pu-
pil tests lower than he was rated, another higher.
Why the discrepancy? In solving such problems, a
good many of which are sure to arise in the testing of
forty pupils, the teacher will gain an insight into the
mentality and character of her children that will richly
repay her for the somewhat difficult task of making
The teacher will find it especially interesting and
instructive to compare her trait ratings with the I Q*s
later found in the tests. By doing so she will see the
dose correlation which usually exists between desira-
ble traits. Notwithstanding occasional exceptions to
the rule, she will find that usually the child she has
rated high in conscientiousness, obedience, will power,
sense of humor, etc., will earn a high I Q in the Binet
test; the child she has rated low, an inferior I Q. 1 In
this way she will come to appreciate the close connec-
tion which often exists between unsatisfactory conduct
and inferior intelligence. In connection with the other
supplementary information the teacher will find it in-
structive to compare the I Q of the various nationali-
ties and occupational groups represented in her class.
The Providence example. The city of Providence,
Bhode Island, offers an excellent illustration of what
1 See Figures 15, 16, and 17, p. 59.
THE USE OF MENTAL TESTS SOS
may be accomplished by training teachers in the use of
mental tests. Under the leadership of Mr. Richard D.
Allen, Director of Vocational Guidance, and of Miss
Greene, Supervisor of Primary Instruction, large num-
bers of the teachers of that city have been taught to
give Binet tests. The instruction is given in a four-
weeks course in the summer normal school, and in-
cludes twenty practice tests. A teachers' dub of two
hundred members has been formed for the purpose of
promoting the grading of school children by mental
ability. Miss Greene's work in testing first-grade chil-
dren has already been mentioned. 1 Mr. Allen has
kindly sent us the following information regarding
I found that at the beginning of my work with the tests
there were a great many puzzling things. For example, I
occasionally found pupils who tested low and were never-
theless doing fair work. In such cases when I took the
mental age into account I usually found that this was above
the mental age of the children with whom they were com-
peting. The facts were then easy to explain. I have yet to
find a single case of the two thousand tests we have made in
which the I Q and mental age do not throw valuable light
upon the reasons for success or failure.
Our tests show that 90 per cent, at least, of school retarda-
tion is without doubt due to mental inferiority. There are
very few seriously retarded children who do not do satis-
factory work in school when they are placed in a grade which
corresponds to mental age. One of the results of placing
children of the same mental age together has been the cut-
ting down of failures by fully 50 per cent.
* Page 64,
304 INTELLIGENCE OP SCHOOL CHILDREN
We have arranged to give an intelligence test to every
child who leaves school to go to work, and we use the test
in determining roughly the limits of the child's vocational
possibilities. We have found, for example, that retarded
boys who drop out of the fifth or sixth grade because of lack
of ability to do the work often succeed well at painting or
plain carpentry. The boys who test higher have, of course,
a wider range of vocational possibilities.
Concerning one group of 1016 children whose men-
tal test and school work had been compared, Mr.
Allen presents the following facts :
1. Of sixty-seven who tested below 70 I Q, sixty-three
made an average school mark of D or E. Of two hun-
dred who tested above 110 1 Q, only four had an average
school mark as low as D.
S. Of the sixty-nine pupils testing below 70, all except
seven were located in a grade above that corresponding
to mental age. Of eighty-four pupils testing above
120, every one was located in a grade below that cor-
responding to mental age. Many were below-grade
as much as three, four, or five years.
8. Of one hundred and three children who are located in a
grade either one and a half or two years above that cor-
responding to mental age, over 90 per cent are failures.
4. A great majority of the children who test under 90
I Q never graduate from the grammar school.
5. There is clearly a very close relation between the place-
ment of a child in school and the quality of work he is
able to perform. Scholarship plus chronological age
plus the grade in which a child is located gives a fairly
good basis for estimating the child's mental age. Con-
versely, the I Q plus the mental age plus the grade
gives a fairly clear estimate of the quality of the work
which the child should be able to do.
THE USE OP MENTAL TESTS 805
Getting the testing done. The earlier in the term
the tests are made the greater their value. Since the
testing must ordinarily be done out of school hours, it is
likely to be two or three months before the teacher can
complete her " pupil survey." One test each afternoon
will dispose of the difficult cases within a few weeks,
and of the entire class in a month or two. Sometimes
Saturdays can be utilized to advantage by making
special appointments with pupils to come to the school
for the purpose. Children invariably like to be tested,
and are always willing to forego an hour of play for the
experience. The teacher will not long regard the work
as an additional burden. The interest in seeing how the
different children respond to the same tests grows to the
point of fascination. The work is also made easier by
noting how the experience adds to the pupil's feeling of
intimacy toward the teacher. To test a child skillfully
nearly always means to win a devoted little friend.
The tests should be made, however, even though
they can only come at the middle or end of the term,
as the results can be used to great advantage in decid-
ing doubtful cases of promotion or double-promotion.
The teacher should record the results in full for each
child in a little book to be kept in her desk for handy
reference. The record should include, after the child's
name, the age (in years and months), the mental age,
the I Q, the nationality of each parent, the occupation
that supports the home, and the various ratings on
school work, intelligence, and other traits. If educa-
306 INTELLIGENCE OP SCHOOL CHILDREN
tional tests have been given, the results of these should
be recorded here also. The teacher who keeps such a
record will soon come to look upon it as indispensable.
The testing should be carried on in such a way as
not to excite undue comment among the pupils. The
teacher will, of course, refrain from speaking of the
tests as " intelligence tests." She may refer to them
merely as " tests to see what children can do." She
can avoid creating apprehension by beginning with the
brightest pupils. She will thus prevent the idea get-
ting abroad that to be given a test means to be sus-
pected of mental inferiority. It is never advisable or
necessary to test a child against his will. After a few
have been given a test the others are invariably anx-
ious to have the same privilege.
The use of abbreviated tests. When possible each
child should be given a complete Binet test, but if
time does not permit this the teacher can make a fairly
satisfactory survey of her pupils by means of an ab-
breviated form of the scale which requires no more
than ten to twenty minutes per pupil, according to
the form of abbreviation used. Although the brief
test falls a good deal short of the complete test in re-
liability, it is far better than nothing.
The following abbreviations of the Stanford Revi-
sion will be found serviceable:
1. The four tests of each year group (six in year 18) indi-
cated in the record booklet by stars. Time required,
approximately 30 minutes.
THE USE OF MENTAL TESTS 307
& Any three tests chosen at random from each year group
(but four in year 12). Time, approximately 20 minutes.
8. The vocabulary test alone. Time, 8 to 10 minutes.
When fewer than the regular number of tests are
used in a year group it is of course necessary to increase
the value of each test in months in proportion to the
reduction of number. In the year groups below 1& ?
each test has a value of months when all six are used
in each year, of 3 months when four tests are used,
and of 4 months when only three tests are used. The
same principle holds in the upper part of the scale.
In year 12, for example, each test has a value of 3
months when all eight are given, of 4 months when
six are given, and of 6 months when four are given.
Perhaps the surest way to avoid errors of weighting
tests is to follow the rule of giving either all the regular
tests or only half of them in each year. If only half
are given, the regular weighting would of course be
doubled and the tests of the different year groups
would have values as follows:
Years I to X 4 months
Year XIE 6 months
Year XIV 8 months
Year XVI 10 months
Year XVHI 12 months
This form of abbreviation can be given to younger
children in fifteen to twenty minutes. Either the first
half of the tests can be given in each year group, or
they can be selected according to the limitations of
308 INTELLIGENCE OF SCHOOL CHILDREN
time or the preference of the examiner. Otis 1 has de-
termined statistically the reliability of either half of
the Stanford-Binet Scale, when it is thus split in two
vertically. His study shows the probable error of an
I Q to be about 4.5 points when half the scale is used
and about 3 points when all of it is used. This means
that in fifty per cent of the cases the I Q found when
half the scale is used would fall within the range of 4|
points above or 4j points below the true I Q; and that
the I Q found when the entire scale is used is in fifty
per cent of cases within the range of 3 points above or
3 points below the true I Q. Half the scale is thus
accurate enough for most practical purposes.
The vocabulary test as a brief intelligence scale.
Where a hasty preliminary sifting of the pupils is
necessary it is recommended that the vocabulary test
be used by itself. It should be given to one child at
a time, taken alone, and requires on an average only
about eight or ten minutes. If the complete Binet
test is given later the vocabulary scores can be added
in and no time will have been lost.
On the basis of the above mental age standards,
the Stanford vocabulary test gives a mental age cor-
1 Otis,' Arthur S., "The Reliability of the Binet Scale," The Pay*
chologicol Clinic, 1919.
THE USE OP MENTAL TESTS 809
rect within one year in about 60 per cent of cases, and
within a year and a half in 80 per cent of cases.
The teacher will doubtless be surprised that any
single test requiring only ten minutes could possess
this degree of accuracy. One might very well sup-
pose that the child's vocabulary would depend upon
home environment and formal instruction, that it
would be an index of special rather than general abil-
ity, and that anyway it could not be accurately enough
measured by a list of 100 words selected at random
from the dictionary. As we have shown elsewhere,
all of these theoretical objections are contradicted by
the facts. 1
That it measures general intelligence rather than
special ability is shown by the high correlation of
vocabulary score with Stanford-Binet mental ages.
Table 41 shows that the correlation for 681 school
children was .91.
The probable error of a mental age based on the
vocabulary score alone is approximately 9| months.
This means that 50 per cent of the vocabulary mental
ages would deviate less than 9j months from the
mental age resulting from a complete Stanford-Binet
test. It would deviate more than 12 months in only
40 per cent of cases and more than 4 months in only
10 per cent of cases.
1 Terman, Lewis M., Kohs, S. C., and others: "The Vocabulary
Test as a Measure of Intelligence," Journal of Educational Psychol-
ogy, October, 1918.
. ,,H .^WOOOOrj^fr-tt
ti-t : : 11*4*110*4 :
THE USE OF MENTAL TESTS 311
The vocabulary is much less influenced by the cul-
tural status of the home than one would expect. The
following illustration is typical: A. B., the feeble-
minded son of a college professor, is 14 years old and
tests at the mental age of 11 by the Stanford-Binet.
His vocabulary score alone indicated a mental age of
11-4. The exceptional language environment of this
boy had raised his vocabulary only a third of a year
above his general mental level. The influence of
exceptionally poor language environment is also very
slight. E. is a Portuguese boy whose parents speak
only broken English. This boy, the brightest we have
tested from a Portuguese family, tested at the mental
age of 18-6 when he was 14-5 chronologically. His
vocabulary score (70 words) was equal to the median
for first-year college students. The extreme poverty
of his language environment had not prevented his
vocabulary from keeping pace with his general level
A vocabulary test of 100 words is sufficient to meas-
ure an individual's total vocabulary very accurately.
When several different word lists of this kind are used
with the same subject they give approximately the
same result. The probable error of a vocabulary
score for a 100 word list is about 2 words, and since
each word in the list represents 180 words in the dic-
tionary, the probable error of total vocabulary based
on the test is 360 words. For example, if a subject
defines 40 words correctly his total vocabulary figures
312 INTELLIGENCE OF SCHOOL CHILDREN
at 40 X ISO, 1 or 7200 words. The chances are 50 to
50 that this subject's actual vocabulary lies within the
range of 7200 plus or minus 360; i.e., between 6840 and
7560. The chances are 22 to 1 that the total vocab-
ulary as calculated from the score in the vocabulary
test will not be found to deviate from the true vocab-
ulary by more than 1000 words.
Group tests. Above the third grade the prelimi-
nary sifting and classification can be done most expe-
ditiously by means of some of the recently devised
group tests. These can be given simultaneously to
all the pupils of a class in fifty to sixty minutes.
Some of the group tests have the great advantage
that they require no extended training either for giv-
ing or scoring. The scoring is done mechanically by
means of stencils, and requires about ten minutes per
pupil. The tests can be given as a regular school
exercise, and the scoring can be done at the teacher's
convenience out of school hours. 2
While no scale has been devised for group testing
1 The Stanford vocabulary list was constructed by selecting every
180th word in a dictionary containing 18,000 words.
* The best scale for group testing available at present is that
of Arthur S. Otis, published by the World Book Company, Yonkers,
New York. Another, which will have a more extensive experimental
basis, is being prepared by an investigating board composed of M. E.
Haggerty, L. M. Terman, E. L. Thorndike, G. M. Whipple, and
R. M. Yerkes. It should be ready for use early in 1920. The in-
vestigation was made possible by an appropriation of $25,000 by
the General Education Board of the Rockefeller Foundation.
For blanks or information address the Section of Psychology,
National Research Council, Washington, D.C.
THE USE OF MENTAL TESTS SIS
which yields as dependable results as the Binet
method, the group tests are deserving of wide vogue.
There should be no rivalry between the group method
and the individual method of testing, as each supple-
ments the other. All the pupils above the third or
fourth grade should be given a group test annually.
We may confidently expect this practice to become
common in the no distant future. The individual
method will find its field in the first three grades,
and in the more thorough examination of children in
the upper grades who make exceptional scores in the
Acceleration, and retardation, 72.
Adenoids, 151, 188.
Age, promotion by, 27, 28, 97,
158; influence on ability to do
school work, 48, 49; of learning
to read, 178; of walking and
talking, 187, 188.
Allen, Richard D., on Binet tests,
Ayres, Leonard P., Laggards in
our Schools, 111.
Backward children, schools for,
42, 43; vocational training for,
Berry, Charles Scott, 114.
Binet, Alfred, 1, 6.
BinetrSimon Intelligence Scale,
the, 2, 5, 12, 25; preparation
needed for use of, 292-302. See
also Stanford-Bine! scale.
Business men, intelligence tests
of, 278, 279.
Campbell, Dr. Macfie, 125.
Class sectioning, 266.
College students, intelligence tests
Comparison, need of absolute
standard of, 70.
Compensation, the theory of, 57.
Cummings, Blanche, 144, 233.
Cuneo, Irene, tests made by, 30,
32, 34, 36, 37, 50, 51, 146.
Curriculum, 29, 32, 90.
Dickson, Virgil E., 42, 119, 296;
tests made by, 43, 44, 47, 48,
52, 55, 56, 57; predicts as to
school progress, 61, 62; on cause
of retardation, 63, 64; striking
result of a mental test by, 289,
Discrepancies between mental
age and school performance,97-
99; tests showing effect of un-
usual application, 99; of child's
personality, 100, 101; of timid-
ity and lack of self-confidence,
102; of mental inertia, 102, 103;
of emotional or nervous tenden-
cies, 103, 104; of "home spoil-
ing," 104, 105; of physical de-
fects, 105; of love affairs and
day dreaming, 105, 106.
Downey, Dr. June E., 280.
Educational guidance, 268, 286-
Educational lock-step, 74.
Elimination, high-school, 86-89.
Environment, effect of, 10-15.
Express company employees, in-
telligence tests of, 275, 276.
Failures, social and industrial, in-
telligence tests of, 282-286.
Feeble-minded, school children,
119; progress of, 121; number
of individuals to be classed as,
124, 127; definition of, 127;
not curable, 128; grade expect-
ancy of, 129-131; a burden,
132; among delinquents, 283,
Fifth grade, individual differ-
ences in, 66-74.
Finger prints, 5.
Firemen of San Jose", intelligence
tests of, 271-274.
First grade, individual differences
in, 42-65; critical importance
of, 42; mental-age differences,
44; mental age necessary for,
45-47; sex differences, 55; racial
and social differences, 55, 56;
feasibility of testing first-grade
children, 64, 65.
Flanders, J. K, 275, 276.
Galton, Frances, 1, 259.
Gaylor, G. W., 114.
Goddard, Dr. H. H., 125.
Group tests, 312, 313.
Gulick, Dr., 115.
Gypsy, Mary, case of, 13.
Haggerty, M. E., 312.
Health, rating on, 185-187.
Heredity, 188-190. See also case
studies of superior children,
High School, individual differ-
ences in first year of, 75-91;
mental-age differences in, 77-
79; mental age and school
marks in, 79, 80.
Hoboes, intelligence tests of, 282-
Hocking, Adeline, 187.
Hollingworth, H. L., 268.
Hollingworth, Leta S., 259.
Home reading, 180, 181.
Hubbard, Margaret Hopwood,
165, 166, 168,
Hubbard, O. S., 66, 175.
Individual differences, amount
and significance of, 17-29;
causes of, in school progress,
21; among kindergarten chil-
dren, 30-41; in the first grade,
42-65; in the fifth grade, 66-74;
in first year of high school, 75-
Intelligence, principles of test-
ing, 1-15; and other traits, 56-
61 ; relation to elimination, 86.
Intelligence quotient, the, 7; a
basis for prediction, 8-10, 70,
157; effect of environment
on, 10-15; distribution of, in
Kindergarten, 32, 33; in first
grade, 50; in High-School pu-
pils, 80; constancy of, 138, 142,
146; of superior children, 170.
Intelligence tests in vocational
and educational guidance, 268-
Irwin, May, 265.
Johnson, Glenn R., 282.
Key, Dr. Wilhelmme, 125.
Kindergarten, difference among
children in, 30-41; sex differ-
ences, 34; girls superior to boys,
35, 36; special need of tests in,
37-39; demand upon intelli-
King, Irving, 88.
Kohs, S. C., 309.
Laggards, school, 111-134.
Learning to read, age of, 178.
Lowery, Howard, 119.
Lurton, F. E., 113.
Mead, C. D., 188.
Mental age, meaning of, 5, 6; a
basis for school grading, 7, 107;
overlapping in grades, 25-27,
80-32; differences in first
grade, 44-47; differences in
High-School pupils, 77; and
school marks, 79, 80, 97;
standard for grading, 92-110;
normal age'for different grades,
93-95; discrepancies with
school performance, 97-106;
objections to grading superior
children by, 262.
Mental deficiency, criteria of,
Mental growth, curves of, 153-
Mental test, significance of, 3, 4;
usefulness of, 109; of school
laggards, 111-134; sugges-
tions for use of, 291-313.
Missing link, the, 4.
Montessori, Madame, 40.
Moral traits, 184, 185.
Morton, W. H. S., 113.
Only children, 177, 178, 231.
Opportunity classes for superior
Otis, Arthur S., 308, 312.
Otis Group Scale, the, 15, 75.
Over-age children, problem of r
28, 29, 134; laggards among,
111, 112, 116; over-promotion
among, 130, 131; special
classes for, 131, 132.
Physical traits, 185-188.
Play and recreation, 179.
Poems and fables by seven-year-
old boy, 204, 205; by boy of
Policeman of San Jose", intelli-
gence tests of, 271-274.
Precocity, not pathological, 165.
Prediction, regarding school
progress, 36, 61-63, 70; the
I Q as a basis for, 135-164; the
essence of science, 135; limita-
tions of, 137, 148.
Preston, Isabel, 92, tests made
by, 98-106; 107, 108.
Proctor, W. M., studies first
year High-School pupils, 75,
Promotion, by age, 27, 28, 97,
158; of over-age children, 130,
Race, Henrietta, 265.
Recreation, 179, 180.
Retardation, prevention of, 21,
28; causes of, 24, 25, 27, 63,
64; and acceleration, 72; chief
cause of, 73, 74, 116, 117;
among school laggards, 1 12; in
rural schools, 114; supposed
causes of, 114-116.
Salesgirls, intelligence tests of,
School grading, failure of, 25, 27;
a model study of, 43-48.
School success, sources of error
in judging, 95.
Sex differences, at kindergarten
age, 34-36; in first grade, 55.
Skipping grades, 177.
Social status, 183.
Specht, Louise F., 265.
Special class, the, 131.
Standard of comparison, neces-
sity of, 70, 71.
Standardization of tests, 3, 4, 5,
Stanford-Binet scale, 1, 2, 5, 10,
25, 50, 66, 75, 76, 98, 107, 167,
168; test of fable interpreta-
tion, 205, 208; Record Book-
let for use with, 293, 294; ab-
breviations of, 306-311.
Strayer, G. D., 87, 112.
Street-car employees, intelli-
gence tests of, 277, 278.
Strong, Dr., Case of Margaret,
Superior children, some facts
about, 165-193; educational
neglect of, 165; case studies of,
194-267; indications of, 260,
261; objections to grading by
mental age, 262; opportunity
classes for, 263-266.
Talent, conservation of, 288-
Talking, age of, 187, 188.
Taussig's division of occupation,
Terman, Lewis M., 119, 134, 146,
187, 195, 203, 209, 259, 271,
291, 297, 309.
Thorndike, E. L., Educational
Psychology, 17 n.
Trait ratings, 181-184.
Vocabulary test, 808-312.
Vocational guidance, intelligence
tests in, 268-290; and educa-
tional guidance inseparable,
268; limitations of, 269, 270.
Vocational training, for back-
ward children, 132.
Walking, age of, 187, 188.
Waugh, C. W., 277.
Whipple, G. M., 266, 312.
Williams, Dr. J. H., 144.
Yerkes, R. M., 312.
1 07 364